id
stringlengths 9
16
| submitter
stringlengths 1
64
⌀ | authors
stringlengths 3
62.8k
| title
stringlengths 1
382
| comments
stringlengths 1
1.15k
⌀ | journal-ref
stringlengths 1
557
⌀ | doi
stringlengths 7
153
⌀ | report-no
stringlengths 1
509
⌀ | categories
stringlengths 5
125
| license
stringclasses 9
values | abstract
stringlengths 3
6.09k
| versions
listlengths 1
187
| update_date
timestamp[s]date 2007-05-23 00:00:00
2025-08-22 00:00:00
| authors_parsed
listlengths 1
3.31k
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2506.03106
|
Xiaoying Zhang
|
Xiaoying Zhang, Hao Sun, Yipeng Zhang, Kaituo Feng, Chaochao Lu, Chao Yang, Helen Meng
|
Critique-GRPO: Advancing LLM Reasoning with Natural Language and Numerical Feedback
|
52 pages, updated with new experimental results and implementation details
| null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recent advances in reinforcement learning (RL) with numerical feedback, such as scalar rewards, have significantly enhanced the complex reasoning capabilities of large language models (LLMs). Despite this success, we identify three key challenges encountered by RL with solely numerical feedback: performance plateaus, limited effectiveness of spontaneous self-reflection, and persistent failures. We then demonstrate that RL-finetuned models, even after exhibiting performance plateaus, can generate correct refinements on persistently failed problems by leveraging natural language feedback in the form of critiques. Building on this insight, we propose Critique-GRPO, an online RL framework that integrates both natural language and numerical feedback for effective policy optimization. Critique-GRPO enables LLMs to learn from initial responses and critique-guided self-refinements simultaneously while maintaining exploration. Additionally, we employ a shaping function to amplify learning from correct, especially unfamiliar, refinements and penalize incorrect ones. Extensive experiments with Qwen2.5-7B-Base, Qwen2.5-Math-7B-Base, and Qwen3-8B demonstrate that Critique-GRPO consistently outperforms supervised learning and RL-based fine-tuning methods across eight challenging mathematical, STEM, and general reasoning tasks. Specifically, Critique-GRPO improves average pass@1 scores across all compared methods by approximately +4.4% on Qwen2.5-7B-Base and +3.8% on Qwen3-8B. Notably, Critique-GRPO enables effective self-improvement through self-critiquing, achieving significant gains over GRPO, e.g., +16.7% pass@1 improvement on AIME 2024.
|
[
{
"version": "v1",
"created": "Tue, 03 Jun 2025 17:39:02 GMT"
},
{
"version": "v2",
"created": "Wed, 04 Jun 2025 13:45:47 GMT"
},
{
"version": "v3",
"created": "Tue, 15 Jul 2025 12:11:37 GMT"
},
{
"version": "v4",
"created": "Thu, 17 Jul 2025 04:08:03 GMT"
},
{
"version": "v5",
"created": "Wed, 20 Aug 2025 09:10:05 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhang",
"Xiaoying",
""
],
[
"Sun",
"Hao",
""
],
[
"Zhang",
"Yipeng",
""
],
[
"Feng",
"Kaituo",
""
],
[
"Lu",
"Chaochao",
""
],
[
"Yang",
"Chao",
""
],
[
"Meng",
"Helen",
""
]
] |
2506.04004
|
Mar\'ia Carranza-Escudero
|
Mar\'ia Carranza-Escudero, Christopher J. Conselice, Nathan Adams, Thomas Harvey, Duncan Austin, Peter Behroozi, Leonardo Ferreira, Katherine Ormerod, Qiao Duan, James Trussler, Qiong Li, Lewi Westcott, Rogier A. Windhorst, Dan Coe, Seth H. Cohen, Cheng Cheng, Simon P. Driver, Brenda Frye, Lukas J. Furtak, Norman A. Grogin, Nimish P. Hathi, Rolf A. Jansen, Anton M. Koekemoer, Madeline A. Marshall, Rosalia O'Brien, Norbert Pirzkal, Maria Polletta, Aaron Robotham, Michael J. Rutkowski, Jake Summers, Stephen M. Wilkins, Christopher N. A. Willmer, Haojing Yan, and Adi Zitrin
|
Lonely Little Red Dots: Challenges to the AGN-nature of little red dots through their clustering and spectral energy distributions
|
24 pages, 17 figures, submitted to ApJL
|
ApJL 989 (2025) L50
|
10.3847/2041-8213/adf73d
| null |
astro-ph.GA
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Observations with the James Webb Space Telescope (JWST) reveal a previously unseen population of compact red objects, known as ``little red dots`` (LRDs). We study a new photometrically selected sample of 124 LRDs in the redshift range $z$ $\sim$ 3 - 10 selected from NIRCam coverage of the CEERS, NEP-TDF, JADES and JEMS surveys. For JADES, the NEP-TDF and CEERS, we compare SED models with and without AGN components and analyse the impact of an AGN component on the goodness of fit using the Bayesian information criterion (BIC). We find that whilst the $\chi^{2}$ of the majority of models containing AGN components is improved compared to models without AGN components, we show that the BIC suggests models without AGN are a more appropriate fit to LRD SEDs, especially when MIRI data is available. We also measure LRD clustering in the CEERS field, JADES field, and NEP-TDF, where we compare the spatial distribution of LRDs and galaxies with Kolmogorov-Smirnov tests of equality of distribution. We find that the neighbourhood of LRDs tends to be less dense compared to galaxies at all selections and masses and at similar redshifts. We further measure upper limit estimates for the halo masses of LRDs using abundance matching. Whilst the population of LRDs could be a mixture of several different inherent populations, as a whole it does appear that these systems are mostly hosting compact galaxies or star clusters in formation.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 14:32:01 GMT"
},
{
"version": "v2",
"created": "Fri, 06 Jun 2025 17:25:37 GMT"
},
{
"version": "v3",
"created": "Wed, 25 Jun 2025 10:01:50 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 16:15:49 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Carranza-Escudero",
"María",
""
],
[
"Conselice",
"Christopher J.",
""
],
[
"Adams",
"Nathan",
""
],
[
"Harvey",
"Thomas",
""
],
[
"Austin",
"Duncan",
""
],
[
"Behroozi",
"Peter",
""
],
[
"Ferreira",
"Leonardo",
""
],
[
"Ormerod",
"Katherine",
""
],
[
"Duan",
"Qiao",
""
],
[
"Trussler",
"James",
""
],
[
"Li",
"Qiong",
""
],
[
"Westcott",
"Lewi",
""
],
[
"Windhorst",
"Rogier A.",
""
],
[
"Coe",
"Dan",
""
],
[
"Cohen",
"Seth H.",
""
],
[
"Cheng",
"Cheng",
""
],
[
"Driver",
"Simon P.",
""
],
[
"Frye",
"Brenda",
""
],
[
"Furtak",
"Lukas J.",
""
],
[
"Grogin",
"Norman A.",
""
],
[
"Hathi",
"Nimish P.",
""
],
[
"Jansen",
"Rolf A.",
""
],
[
"Koekemoer",
"Anton M.",
""
],
[
"Marshall",
"Madeline A.",
""
],
[
"O'Brien",
"Rosalia",
""
],
[
"Pirzkal",
"Norbert",
""
],
[
"Polletta",
"Maria",
""
],
[
"Robotham",
"Aaron",
""
],
[
"Rutkowski",
"Michael J.",
""
],
[
"Summers",
"Jake",
""
],
[
"Wilkins",
"Stephen M.",
""
],
[
"Willmer",
"Christopher N. A.",
""
],
[
"Yan",
"Haojing",
""
],
[
"Zitrin",
"Adi",
""
]
] |
2506.04057
|
Pinrui Shen
|
Pinrui Shen and Mariusz Pawlak and Donald Booth and Kent Nickerson and Haddad Miladi and H. R. Sadeghpour and James P. Shaffer
|
Precision measurement of Cs($nF_J$) quantum defects and calculations of scalar and tensor polarizabilities of the $nS_{1/2}$, $nP_J$ ,$nD_J$ , and $nF_J$ series
|
13 pages, 7 figures, 11 tables
|
Phys. Rev. A 112, 022814 (2025)
|
10.1103/dkb4-kb5b
| null |
physics.atom-ph quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we extend our recent work on cesium S and D states [Phys. Rev. Lett. 133, 233005 (2024)] to the F states. We present absolute frequency measurements of the $|6S_{1/2}, F = 3\rangle \rightarrow nF_{5/2,7/2}(n = 28-68)$ Rydberg series to measure the spectrum of $^{133}$Cs. Atomic spectra are obtained using a three-photon excitation scheme referenced to an optical frequency comb in a sample of ultracold $^{133}$Cs. By globally fitting the absolute-frequency measurements to the modified Ritz formula, we determine the quantum defects of the $nF_{5/2}$ and $nF_{7/2}$ series. The ionization potential extracted for both series from the modified Ritz formula agrees with our measurements based on the S and D series. Fine-structure intervals are calculated and parameterized. The wave functions computed for the energies from the quantum defects are used to calculate transition dipole moments. We compare the reduced electric-dipole matrix elements with available benchmarks and find agreement within the precision of those works. The scalar and tensor polarizabilities of the $nS_{1/2}$, $nP_J$ , $nD_J$ and $nF_J$ series are calculated based on the now more accurate set of wave functions. Moreover, we report the polarizability as a series in powers of the effective principal quantum number and find the main coefficients of the expansion. The results will be useful for calculating properties of $^{133}$Cs such as collision and decay rates, polarizabilities, and magic wavelengths.
|
[
{
"version": "v1",
"created": "Wed, 04 Jun 2025 15:21:51 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 01:22:00 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Shen",
"Pinrui",
""
],
[
"Pawlak",
"Mariusz",
""
],
[
"Booth",
"Donald",
""
],
[
"Nickerson",
"Kent",
""
],
[
"Miladi",
"Haddad",
""
],
[
"Sadeghpour",
"H. R.",
""
],
[
"Shaffer",
"James P.",
""
]
] |
2506.04236
|
Botao Amber Hu
|
Botao Amber Hu, Helena Rong
|
Spore in the Wild: A Case Study of Spore.fun as an Open-Environment Evolution Experiment with Sovereign AI Agents on TEE-Secured Blockchains
|
Accepted by ALIFE 2025
| null | null | null |
cs.MA cs.AI cs.HC cs.NE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
In Artificial Life (ALife) research, replicating Open-Ended Evolution (OEE)-the continuous emergence of novelty observed in biological life-has usually been pursued within isolated, closed system simulations, such as Tierra and Avida, which have typically plateaued after an initial burst of novelty, failing to achieve sustained OEE. Scholars suggest that OEE requires an open-environment system that continually exchanges information or energy with its environment. A recent technological innovation in Decentralized Physical Infrastructure Network (DePIN), which provides permissionless computational substrates, enables the deployment of Large Language Model-based AI agents on blockchains integrated with Trusted Execution Environments (TEEs). This enables on-chain agents to operate autonomously "in the wild," achieving self-sovereignty without human oversight. These agents can control their own social media accounts and cryptocurrency wallets, allowing them to interact directly with blockchain-based financial networks and broader human social media. Building on this new paradigm of on-chain agents, Spore.fun is a recent real-world AI evolution experiment that enables autonomous breeding and evolution of new on-chain agents. This paper presents a detailed case study of Spore.fun, examining agent behaviors and their evolutionary trajectories through digital ethology. We aim to spark discussion about whether open-environment ALife systems "in the wild," based on permissionless computational substrates and driven by economic incentives to interact with their environment, could finally achieve the long-sought goal of OEE.
|
[
{
"version": "v1",
"created": "Sat, 24 May 2025 14:42:36 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 20:40:20 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hu",
"Botao Amber",
""
],
[
"Rong",
"Helena",
""
]
] |
2506.05187
|
EPTCS
|
Alastair A. Abbott, Mehdi Mhalla, Pierre Pocreau
|
Classical and Quantum Query Complexity of Boolean Functions under Indefinite Causal Order
|
In Proceedings QPL 2025, arXiv:2508.13619
|
EPTCS 426, 2025, pp. 270-300
|
10.4204/EPTCS.426.11
| null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Computational models typically assume that operations are applied in a fixed sequential order. In recent years several works have looked at relaxing this assumption, considering computations without any fixed causal structure and showing that such ''causally indefinite'' computations can provide advantages in various tasks. Recently, the quantum query complexity of Boolean functions has been used as a tool to probe their computational power in a standard complexity theoretic framework, but no separation in exact query complexity has thus-far been found. In this paper, we investigate this problem starting with the simpler and fully classical notion of deterministic query complexity of Boolean functions, and using classical-deterministic processes -- which may exhibit causal indefiniteness -- as a generalised computational framework. We first show that the standard polynomial and certificate lower bounds of deterministic query complexity also hold in such generalised models. Then, we formulate a Boolean function for which causal indefiniteness permits a reduction in query complexity and show that this advantage can be amplified into a polynomial separation. Finally, with the insights gained in the classical-deterministic setting, we give a Boolean function whose quantum query complexity is reduced by causally indefinite computations.
|
[
{
"version": "v1",
"created": "Thu, 05 Jun 2025 16:00:36 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:45:08 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Abbott",
"Alastair A.",
""
],
[
"Mhalla",
"Mehdi",
""
],
[
"Pocreau",
"Pierre",
""
]
] |
2506.05730
|
Mike Steel Prof.
|
Qiang Zhang and Mike Steel
|
Counting rankings of tree-child networks
|
15 pages,6 figures
| null | null | null |
q-bio.PE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Rooted phylogenetic networks allow biologists to represent evolutionary relationships between present-day species by revealing ancestral speciation and hybridization events. A convenient and well-studied class of such networks are `tree-child networks' and a `ranking' of such a network is a temporal ordering of the ancestral speciation and hybridization events. In this short note, we investigate the question of counting such rankings on any given binary (or semi-binary) tree-child network. We also consider a class of binary tree-child networks that have exactly one ranking, and investigate further the relationship between ranked-tree child networks and the class of `normal' networks. Finally, we provide an explicit asymptotic expression for the expected number of rankings of a tree-child network chosen uniformly at random.
|
[
{
"version": "v1",
"created": "Fri, 06 Jun 2025 04:24:23 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 01:38:40 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhang",
"Qiang",
""
],
[
"Steel",
"Mike",
""
]
] |
2506.06766
|
Bixiang Wang
|
Bixiang Wang
|
Well-posedness of Fractional Stochastic p-Laplace Equations Driven by Superlinear Transport Noise
| null | null | null | null |
math.PR math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we prove the existence and uniqueness of solutions of the fractional p-Laplace equation with a polynomial drift of arbitrary order driven by superlinear transport noise. By the monotone argument, we first prove the existence and uniqueness of solutions of an abstract stochastic differential equation satisfying a fully local monotonicity condition. We then apply the abstract result to the fractional stochastic p-Laplace equation defined in a bounded domain. The main difficulty is to establish the tightness as well as the uniform integrability of a sequence of approximate solutions defined by the Galerkin method. To obtain the necessary uniform estimates, we employ the Skorokhod-Jakubowski representation theorem on a topological space instead of a metric space. Since the strong Skorokhod representation theorem is incorrect even in a complete separable metric space, we pass to the limit of stochastic integrals with respect to a sequence of Wiener processes by a weak convergence argument.
|
[
{
"version": "v1",
"created": "Sat, 07 Jun 2025 11:18:03 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 21:01:11 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wang",
"Bixiang",
""
]
] |
2506.06978
|
Zitian Li
|
Zitian Li and Wang Chi Cheung
|
Near Optimal Non-asymptotic Sample Complexity of 1-Identification
| null | null | null | null |
cs.LG stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Motivated by an open direction in existing literature, we study the 1-identification problem, a fundamental multi-armed bandit formulation on pure exploration. The goal is to determine whether there exists an arm whose mean reward is at least a known threshold $\mu_0$, or to output None if it believes such an arm does not exist. The agent needs to guarantee its output is correct with probability at least $1-\delta$. Degenne & Koolen 2019 has established the asymptotically tight sample complexity for the 1-identification problem, but they commented that the non-asymptotic analysis remains unclear. We design a new algorithm Sequential-Exploration-Exploitation (SEE), and conduct theoretical analysis from the non-asymptotic perspective. Novel to the literature, we achieve near optimality, in the sense of matching upper and lower bounds on the pulling complexity. The gap between the upper and lower bounds is up to a polynomial logarithmic factor. The numerical result also indicates the effectiveness of our algorithm, compared to existing benchmarks.
|
[
{
"version": "v1",
"created": "Sun, 08 Jun 2025 03:23:45 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:44:30 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Li",
"Zitian",
""
],
[
"Cheung",
"Wang Chi",
""
]
] |
2506.07129
|
Jingze Ding
|
Jingze Ding, Zijian Zhou, Lipeng Zhu, Yuping Zhao, Bingli Jiao, Rui Zhang
|
Energy Efficiency Maximization for Movable Antenna Communication Systems
| null | null |
10.1109/TWC.2025.3597735
| null |
cs.IT eess.SP math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
This paper investigates energy efficiency maximization for movable antenna (MA)-aided multi-user uplink communication systems by considering the time delay and energy consumption incurred by practical antenna movement. We first examine the special case with a single user and propose an optimization algorithm based on the one-dimensional (1D) exhaustive search to maximize the user's energy efficiency. Moreover, we derive an upper bound on the energy efficiency and analyze the conditions required to achieve this performance bound under different numbers of channel paths. Then, for the general multi-user scenario, we propose an iterative algorithm to fairly maximize the minimum energy efficiency among all users. Simulation results demonstrate the effectiveness of the proposed scheme in improving energy efficiency compared to existing MA schemes that do not account for movement-related costs, as well as the conventional fixed-position antenna (FPA) scheme. In addition, the results show the robustness of the proposed scheme to imperfect channel state information (CSI) and provide valuable insights for practical system deployment.
|
[
{
"version": "v1",
"created": "Sun, 08 Jun 2025 13:15:34 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ding",
"Jingze",
""
],
[
"Zhou",
"Zijian",
""
],
[
"Zhu",
"Lipeng",
""
],
[
"Zhao",
"Yuping",
""
],
[
"Jiao",
"Bingli",
""
],
[
"Zhang",
"Rui",
""
]
] |
2506.07175
|
Izaak Neri
|
Izaak Neri
|
Thermodynamic Cost of Random-Time Protocols
|
16 pages, 9 figures
| null | null | null |
cond-mat.stat-mech
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Systems that are driven by a randomly timed, external protocol can seemingly violate the second law of thermodynamics. We show that this thermodynamic paradox is resolved if the outcome of the random time is stored in a memory device. Specifically, we show that the average work required to erase the memory is always larger than the average work gained from the protocol. We also discuss experimentally implementable setups that measure random times directly without continuous monitoring. Taken together, this paper discusses the relationship between temporal information and thermodynamics. This framework is relevant for external protocols employing random times, such as, stochastic resetting protocols and cyclically driven heat engines that use randomly timed protocols.
|
[
{
"version": "v1",
"created": "Sun, 08 Jun 2025 14:50:25 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 22:32:06 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Neri",
"Izaak",
""
]
] |
2506.07614
|
Rishikesh Srinivasan
|
Rishikesh Srinivasan, Dheeraj Nagaraj
|
Poisson Midpoint Method for Log Concave Sampling: Beyond the Strong Error Lower Bounds
| null | null | null | null |
math.PR cs.LG math.ST stat.TH
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the problem of sampling from strongly log-concave distributions over $\mathbb{R}^d$ using the Poisson midpoint discretization (a variant of the randomized midpoint method) for overdamped/underdamped Langevin dynamics. We prove its convergence in the 2-Wasserstein distance ($W_2$), achieving a cubic speedup in dependence on the target accuracy ($\epsilon$) over the Euler-Maruyama discretization, surpassing existing bounds for randomized midpoint methods. Notably, in the case of underdamped Langevin dynamics, we demonstrate the complexity of $W_2$ convergence is much smaller than the complexity lower bounds for convergence in $L^2$ strong error established in the literature.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 10:27:15 GMT"
},
{
"version": "v2",
"created": "Mon, 14 Jul 2025 16:39:03 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 09:06:53 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Srinivasan",
"Rishikesh",
""
],
[
"Nagaraj",
"Dheeraj",
""
]
] |
2506.08113
|
Timoth\'ee Hornek
|
Timoth\'ee Hornek Amir Sartipi and Igor Tchappi and Gilbert Fridgen
|
Benchmarking Pre-Trained Time Series Models for Electricity Price Forecasting
| null | null |
10.1109/eem64765.2025.11050326
| null |
cs.LG cs.AI q-fin.ST
|
http://creativecommons.org/licenses/by/4.0/
|
Accurate electricity price forecasting (EPF) is crucial for effective decision-making in power trading on the spot market. While recent advances in generative artificial intelligence (GenAI) and pre-trained large language models (LLMs) have inspired the development of numerous time series foundation models (TSFMs) for time series forecasting, their effectiveness in EPF remains uncertain. To address this gap, we benchmark several state-of-the-art pretrained models--Chronos-Bolt, Chronos-T5, TimesFM, Moirai, Time-MoE, and TimeGPT--against established statistical and machine learning (ML) methods for EPF. Using 2024 day-ahead auction (DAA) electricity prices from Germany, France, the Netherlands, Austria, and Belgium, we generate daily forecasts with a one-day horizon. Chronos-Bolt and Time-MoE emerge as the strongest among the TSFMs, performing on par with traditional models. However, the biseasonal MSTL model, which captures daily and weekly seasonality, stands out for its consistent performance across countries and evaluation metrics, with no TSFM statistically outperforming it.
|
[
{
"version": "v1",
"created": "Mon, 09 Jun 2025 18:10:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:59:08 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sartipi",
"Timothée Hornek Amir",
""
],
[
"Tchappi",
"Igor",
""
],
[
"Fridgen",
"Gilbert",
""
]
] |
2506.08659
|
Ayaka Shimizu
|
Yuko Ozawa, Ayaka Shimizu, Yoshiro Yaguchi
|
The CN matrix of a pure braid projection
|
31 pages, 28 figures
| null | null | null |
math.GT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The CN matrix of an $n$-braid projection $B$ is an $n \times n$ matrix such that each $(i,j)$ entry indicates the number of crossings between $i^{th}$ and $j^{th}$ strands of $B$. In this paper, several patterns of an $n \times n$ matrix to be a CN matrix are discussed, and the CN matrix of a pure 6-braid projection is characterized. As an application, the OU matrix of a pure 6-braid diagram and the crossing matrix of a positive pure 6-braid are also characterized.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 10:20:09 GMT"
},
{
"version": "v2",
"created": "Sat, 19 Jul 2025 05:27:36 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 07:49:57 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ozawa",
"Yuko",
""
],
[
"Shimizu",
"Ayaka",
""
],
[
"Yaguchi",
"Yoshiro",
""
]
] |
2506.08810
|
Carla Groenland
|
Marthe Bonamy and Carla Groenland and Tom Johnston and Natasha Morrison and Alex Scott
|
Infinite induced-saturated graphs
|
26 pages, 13 figures
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A graph $G$ is $H$-induced-saturated if $G$ is $H$-free but deleting any edge or adding any edge creates an induced copy of $H$. There are non-trivial graphs $H$, such as $P_4$, for which no finite $H$-induced-saturated graph $G$ exists. We show that for every finite graph $H$ that is not a clique or an independent set, there always exists a countable $H$-induced-saturated graph. In fact, we show that a far stronger property can be achieved: there is a countably infinite $H$-free graph $G$ such that any graph $G'\ne G$ obtained by making a locally finite set of changes to $G$ contains a copy of $H$.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 14:00:59 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 10:44:56 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bonamy",
"Marthe",
""
],
[
"Groenland",
"Carla",
""
],
[
"Johnston",
"Tom",
""
],
[
"Morrison",
"Natasha",
""
],
[
"Scott",
"Alex",
""
]
] |
2506.09043
|
Tianyu Chen
|
Tianyu Chen, Darshal Shetty, Jeremy G. Siek, Chao-Hong Chen, Weixi Ma, Arnaud Venet, Rocky Liu
|
Gradual Metaprogramming
|
14 pages, 10 figures
| null |
10.1145/3759538.3759650
| null |
cs.PL
|
http://creativecommons.org/licenses/by/4.0/
|
Data engineers increasingly use domain-specific languages (DSLs) to generate the code for data pipelines. Such DSLs are often embedded in Python. Unfortunately, there are challenges in debugging the generation of data pipelines: an error in a Python DSL script is often detected too late, after the execution of the script, and the source code location that triggers the error is hard to pinpoint.
In this paper, we focus on the scenario where a DSL embedded in Python (so it is dynamically-typed) generates data pipeline description code that is statically-typed. We propose gradual metaprogramming to (1) provide a migration path toward statically typed DSLs, (2) immediately provide earlier detection of code generation type errors, and (3) report the source code location responsible for the type error. Gradual metaprogramming accomplishes this by type checking code fragments and incrementally performing runtime checks as they are spliced together. We define MetaGTLC, a metaprogramming calculus in which a gradually-typed metalanguage manipulates a statically-typed object language, and give semantics to it by translation to the cast calculus MetaCC. We prove that successful metaevaluation always generates a well-typed object program and mechanize the proof in Agda.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 17:58:32 GMT"
},
{
"version": "v2",
"created": "Mon, 16 Jun 2025 14:40:41 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 13:01:01 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chen",
"Tianyu",
""
],
[
"Shetty",
"Darshal",
""
],
[
"Siek",
"Jeremy G.",
""
],
[
"Chen",
"Chao-Hong",
""
],
[
"Ma",
"Weixi",
""
],
[
"Venet",
"Arnaud",
""
],
[
"Liu",
"Rocky",
""
]
] |
2506.09117
|
Adam Dillamore
|
Adam M. Dillamore and Jason L. Sanders
|
Bar-driven dispersal of Galactic substructure
|
17 pages, 12 figures, published in MNRAS
| null |
10.1093/mnras/staf1264
| null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Galactic archaeologists often assume that integrals of motion (IoMs) such as $L_z$ and $E$ are conserved, so substructure remains frozen in IoM space over many Gyr. However, this is not true in the Milky Way due in part to its rotating bar. In this study we quantify the effects of the bar on the dynamics of substructure. We employ three different theoretical models: an analytical toy model; a set of test particle simulations with steady and slowing bars; and a cosmological zoom-in simulation of a Milky Way-like galaxy. Each model predicts that the bar increases the angular momentum and energy spread of low-energy substructures by a factor of $\sim10-100$, so they cannot remain tightly clustered. We derive a criterion for determining when this effect is important. The most affected orbits are low energy ($E\lesssim E_\odot$, $r_\mathrm{apo}<40$ kpc), prograde, eccentric, or low inclination. This includes $\sim3/4$ of Galactic globular clusters and $\sim1/4$ of known stellar streams. We predict the presence of abundant bar-dispersed substructure. The structures remain much more tightly clustered in the space of metallicity and Jacobi integral $H_\mathrm{J}=E-\Omega_\mathrm{b}L_z$. We therefore propose using $H_\mathrm{J}$ and chemistry instead of traditional IoMs when searching for inner halo substructure. In $(L_z,E)$ space the dispersal of the structures is along a principal direction with gradient $\mathrm{d}E/\mathrm{d}L_z$ equal to the bar's pattern speed $\Omega_\mathrm{b}$. Bar-dispersed substructure should therefore allow the past evolution of $\Omega_\mathrm{b}$ to be constrained.
|
[
{
"version": "v1",
"created": "Tue, 10 Jun 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 11:21:58 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Dillamore",
"Adam M.",
""
],
[
"Sanders",
"Jason L.",
""
]
] |
2506.09482
|
Dingcheng Zhen
|
Dingcheng Zhen, Qian Qiao, Xu Zheng, Tan Yu, Kangxi Wu, Ziwei Zhang, Siyuan Liu, Shunshun Yin, Ming Tao
|
Marrying Autoregressive Transformer and Diffusion with Multi-Reference Autoregression
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce TransDiff, the first image generation model that marries Autoregressive (AR) Transformer with diffusion models. In this joint modeling framework, TransDiff encodes labels and images into high-level semantic features and employs a diffusion model to estimate the distribution of image samples. On the ImageNet 256x256 benchmark, TransDiff significantly outperforms other image generation models based on standalone AR Transformer or diffusion models. Specifically, TransDiff achieves a Frechet Inception Distance (FID) of 1.61 and an Inception Score (IS) of 293.4, and further provides x2 faster inference latency compared to state-of-the-art methods based on AR Transformer and x112 faster inference compared to diffusion-only models. Furthermore, building on the TransDiff model, we introduce a novel image generation paradigm called Multi-Reference Autoregression (MRAR), which performs autoregressive generation by predicting the next image. MRAR enables the model to reference multiple previously generated images, thereby facilitating the learning of more diverse representations and improving the quality of generated images in subsequent iterations. By applying MRAR, the performance of TransDiff is improved, with the FID reduced from 1.61 to 1.42. We expect TransDiff to open up a new frontier in the field of image generation.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 07:50:31 GMT"
},
{
"version": "v2",
"created": "Sun, 15 Jun 2025 05:03:17 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 06:57:17 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhen",
"Dingcheng",
""
],
[
"Qiao",
"Qian",
""
],
[
"Zheng",
"Xu",
""
],
[
"Yu",
"Tan",
""
],
[
"Wu",
"Kangxi",
""
],
[
"Zhang",
"Ziwei",
""
],
[
"Liu",
"Siyuan",
""
],
[
"Yin",
"Shunshun",
""
],
[
"Tao",
"Ming",
""
]
] |
2506.09733
|
Minjong Cheon
|
Minjong Cheon
|
AtmosMJ: Revisiting Gating Mechanism for AI Weather Forecasting Beyond the Year Scale
|
All authors of this manuscript have not reached a consensus on its submission to arXiv. Since at least one co-author does not agree with the current version being publicly available, we respectfully request the withdrawal of this preprint in accordance with the authors' collective decision
| null | null | null |
cs.LG cs.AI cs.CV physics.ao-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The advent of Large Weather Models (LWMs) has marked a turning point in data-driven forecasting, with many models now outperforming traditional numerical systems in the medium range. However, achieving stable, long-range autoregressive forecasts beyond a few weeks remains a significant challenge. Prevailing state-of-the-art models that achieve year-long stability, such as SFNO and DLWP-HPX, have relied on transforming input data onto non-standard spatial domains like spherical harmonics or HEALPix meshes. This has led to the prevailing assumption that such representations are necessary to enforce physical consistency and long-term stability. This paper challenges that assumption by investigating whether comparable long-range performance can be achieved on the standard latitude-longitude grid. We introduce AtmosMJ, a deep convolutional network that operates directly on ERA5 data without any spherical remapping. The model's stability is enabled by a novel Gated Residual Fusion (GRF) mechanism, which adaptively moderates feature updates to prevent error accumulation over long recursive simulations. Our results demonstrate that AtmosMJ produces stable and physically plausible forecasts for about 500 days. In quantitative evaluations, it achieves competitive 10-day forecast accuracy against models like Pangu-Weather and GraphCast, all while requiring a remarkably low training budget of 5.7 days on a V100 GPU. Our findings suggest that efficient architectural design, rather than non-standard data representation, can be the key to unlocking stable and computationally efficient long-range weather prediction.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 13:38:56 GMT"
},
{
"version": "v2",
"created": "Wed, 06 Aug 2025 03:40:48 GMT"
},
{
"version": "v3",
"created": "Mon, 18 Aug 2025 02:03:00 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Cheon",
"Minjong",
""
]
] |
2506.10780
|
Edward F. Valeev
|
Samuel R. Powell, Kshitijkumar A. Surjuse, Bimal Gaudel, Edward F. Valeev
|
Slimmer Geminals For Accurate F12 Electronic Structure Models
| null | null | null | null |
physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Slater-type F12 geminal lengthscales originally tuned for the second-order M{\o}ller-Plesset F12 method are too large for higher-order F12 methods formulated using the SP (diagonal fixed-coefficient spin-adapted) F12 ansatz. The new geminal parameters reported herein reduce the basis set incompleteness errors (BSIE) of absolute coupled-cluster singles and doubles F12 correlation energies by a significant - and increasing with the cardinal number of the basis - margin. The effect of geminal reoptimization is especially pronounced for the cc-pVXZ-F12 basis sets (specifically designed for use with F12 methods) relative to their conventional aug-cc-pVXZ counterparts. The BSIEs of relative energies are less affected but substantial reductions can be obtained, especially for atomization energies and ionization potentials with the cc-pVXZ-F12 basis sets. The new geminal parameters are therefore recommended for all applications of high-order F12 methods, such as the coupled-cluster F12 methods and the transcorrelated F12 methods.
|
[
{
"version": "v1",
"created": "Thu, 12 Jun 2025 14:54:04 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 15:55:07 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Powell",
"Samuel R.",
""
],
[
"Surjuse",
"Kshitijkumar A.",
""
],
[
"Gaudel",
"Bimal",
""
],
[
"Valeev",
"Edward F.",
""
]
] |
2506.11091
|
Shaoshi Ling
|
Shaoshi Ling, Guoli Ye
|
Customizing Speech Recognition Model with Large Language Model Feedback
| null | null | null | null |
cs.CL cs.SD eess.AS
|
http://creativecommons.org/licenses/by/4.0/
|
Automatic speech recognition (ASR) systems have achieved strong performance on general transcription tasks. However, they continue to struggle with recognizing rare named entities and adapting to domain mismatches. In contrast, large language models (LLMs), trained on massive internet-scale datasets, are often more effective across a wide range of domains. In this work, we propose a reinforcement learning based approach for unsupervised domain adaptation, leveraging unlabeled data to enhance transcription quality, particularly the named entities affected by domain mismatch, through feedback from a LLM. Given contextual information, our framework employs a LLM as the reward model to score the hypotheses from the ASR model. These scores serve as reward signals to fine-tune the ASR model via reinforcement learning. Our method achieves a 21\% improvement on entity word error rate over conventional self-training methods.
|
[
{
"version": "v1",
"created": "Thu, 05 Jun 2025 18:42:57 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 20:44:16 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ling",
"Shaoshi",
""
],
[
"Ye",
"Guoli",
""
]
] |
2506.11414
|
Zhongtian Hu
|
Zhongtian Hu
|
Exponential Vorticity Hessian Growth in Capillary Liquid Drop in Two Dimensions
|
16 pages, 1 figure; updated reference and contact information
| null | null | null |
math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we concern ourselves with the evolution of a droplet of an ideal fluid in two dimensions, which has nontrivial bulk vorticity and is only subject to the effects of surface tension. We construct initial data with initial domain being a disk and initial velocity being arbitrarily small, such that the vorticity Hessian grows exponentially infinitely in time.
|
[
{
"version": "v1",
"created": "Fri, 13 Jun 2025 02:24:27 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:20:28 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hu",
"Zhongtian",
""
]
] |
2506.13257
|
Tibor Szendrei
|
David Kohns, Tibor Szendrei
|
Joint Quantile Shrinkage: A State-Space Approach toward Non-Crossing Bayesian Quantile Models
| null | null | null | null |
stat.ME econ.EM math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
Crossing of fitted conditional quantiles is a prevalent problem for quantile regression models. We propose a new Bayesian modelling framework that penalises multiple quantile regression functions toward the desired non-crossing space. We achieve this by estimating multiple quantiles jointly with a prior on variation across quantiles, a fused shrinkage prior with quantile adaptivity. The posterior is derived from a decision-theoretic general Bayes perspective, whose form yields a natural state-space interpretation aligned with Time-Varying Parameter (TVP) models. Taken together our approach leads to a Quantile-Varying Parameter (QVP) model, for which we develop efficient sampling algorithms. We demonstrate that our proposed modelling framework provides superior parameter recovery and predictive performance compared to competing Bayesian and frequentist quantile regression estimators in simulated experiments and a real-data application to multivariate quantile estimation in macroeconomics.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 08:56:21 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 06:09:47 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kohns",
"David",
""
],
[
"Szendrei",
"Tibor",
""
]
] |
2506.13636
|
Kerwann Tep
|
Kerwann Tep, Brian T. Cook, Carl L. Rodriguez, Jiya Jolly, Eddie Sawin, Michael S. Petersen, Christoph Gaffud
|
KRIOS: A new basis-expansion $N$-body code for collisional stellar dynamics
|
23 pages, 15 figures, submitted to ApJ
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
The gravitational $N$-body problem is a nearly universal problem in astrophysics which, despite its deceptive simplicity, still presents a significant computational challenge. For collisional systems such as dense star clusters, the need to resolve individual encounters between $N$ stars makes the direct summation of forces - with quadratic complexity - almost infeasible for systems with $N\gtrsim 10^6$ particles over many relaxation times. At the same time, the most common Monte Carlo $N$-body algorithm - that of H\'enon - assumes the cluster to be spherically symmetric. This greatly limits the study of many important features of star clusters, including triaxiality, rotation, and the production of tidal debris. In this paper, we present a new hybrid code, KRIOS, that combines 3D collisionless relaxation using an adaptive self-consistent field method with collisional dynamics handled via H\'enon's method. We demonstrate that KRIOS can accurately model the long-term evolution of clusters and provide its complete phase-space information over many relaxation times. As a test of our new code, we present detailed comparisons to well-known results from stellar dynamics: (i) the collisional evolution of a family of Plummer spheres with varying anisotropy and rotation to core collapse, and (ii) the emergence of the radial-orbit instability in radially anisotropic star clusters, including its non-spherical effects.
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 16:01:02 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 15:56:05 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tep",
"Kerwann",
""
],
[
"Cook",
"Brian T.",
""
],
[
"Rodriguez",
"Carl L.",
""
],
[
"Jolly",
"Jiya",
""
],
[
"Sawin",
"Eddie",
""
],
[
"Petersen",
"Michael S.",
""
],
[
"Gaffud",
"Christoph",
""
]
] |
2506.13790
|
Tapio Pitk\"aranta Mr
|
Tapio Pitk\"aranta
|
The NordDRG AI Benchmark for Large Language Models
|
23 pages, 4 figures
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) are being piloted for clinical coding and decision support, yet no open benchmark targets the hospital-funding layer where Diagnosis-Related Groups (DRGs) determine reimbursement. In most OECD systems, DRGs route a substantial share of multi-trillion-dollar health spending through governed grouper software, making transparency and auditability first-order concerns. We release NordDRG-AI-Benchmark, the first public, rule-complete test bed for DRG reasoning. The package includes (i) machine-readable approximately 20-sheet NordDRG definition tables and (ii) expert manuals and change-log templates that capture governance workflows. It exposes two suites: a 13-task Logic benchmark (code lookup, cross-table inference, grouping features, multilingual terminology, and CC/MCC validity checks) and a 13-task Grouper benchmark that requires full DRG grouper emulation with strict exact-match scoring on both the DRG and the triggering drg_logic.id. Lightweight reference agents (LogicAgent, GrouperAgent) enable artefact-only evaluation. Under an artefact-only (no web) setting, on the 13 Logic tasks GPT-5 Thinking and Opus 4.1 score 13/13, o3 scores 12/13; mid-tier models (GPT-5 Thinking Mini, o4-mini, GPT-5 Fast) achieve 6-8/13, and remaining models score 5/13 or below. On full grouper emulation across 13 tasks, GPT-5 Thinking solves 7/13, o3 6/13, o4-mini 3/13; GPT-5 Thinking Mini solves 1/13, and all other tested endpoints score 0/13. To our knowledge, this is the first public report of an LLM partially emulating the complete NordDRG grouper logic with governance-grade traceability. Coupling a rule-complete release with exact-match tasks and open scoring provides a reproducible yardstick for head-to-head and longitudinal evaluation in hospital funding. Benchmark materials available in Github.
|
[
{
"version": "v1",
"created": "Wed, 11 Jun 2025 11:40:11 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 01:54:22 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 13:47:47 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Pitkäranta",
"Tapio",
""
]
] |
2506.14060
|
Taras Banakh
|
Taras Banakh
|
Linear Geometry and Algebra
|
45 chapters, 802 pages
| null | null | null |
math.AG math.AC math.GR math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Linear Geometry studies geometric properties which can be expressed via the notion of a line. All information about lines is encoded in a ternary relation called a line relation. A set endowed with a line relation is called a liner. So, Linear Geometry studies liners. Imposing some additional axioms on a liner, we obtain some special classes of liners: regular, projective, affine, proaffine, etc. Linear Geometry includes Affine and Projective Geometries and is a part of Incidence Geometry. The aim of this book is to present a self-contained logical development of Linear Geometry, starting with some intuitive acceptable geometric axioms and ending with algebraic structures that necessarily arise from studying the structure of geometric objects that satisfy those simple and intuitive geometric axioms. We shall meet many quite exotic algebraic structures that arise this way: magmas, loops, ternars, quasi-fields, alternative rings, procorps, profields, etc. We strongly prefer (synthetic) geometric proofs and use tools of analytic geometry only when no purely geometric proof is available. Liner Geometry has been developed by many great mathematicians since times of Antiquity (Thales, Euclides, Proclus, Pappus), through Renaissance (Descartes, Desargues), Early Modernity (Playfair, Gauss, Lobachevski, Bolyai, Poncelet, Steiner, M\"obius), Late Modernity Times (Steinitz, Klein, Hilbert, Moufang, Hessenberg, Jordan, Beltrami, Fano, Gallucci, Veblen, Wedderburn, Lenz, Barlotti) till our contempories (Hartshorne, Hall, Buekenhout, Gleason, Kantor, Doyen, Hubault, Dembowski, Klingenberg, Grundh\"ofer).
|
[
{
"version": "v1",
"created": "Mon, 16 Jun 2025 23:44:20 GMT"
},
{
"version": "v2",
"created": "Wed, 18 Jun 2025 20:07:07 GMT"
},
{
"version": "v3",
"created": "Sat, 28 Jun 2025 05:03:42 GMT"
},
{
"version": "v4",
"created": "Mon, 21 Jul 2025 15:06:52 GMT"
},
{
"version": "v5",
"created": "Tue, 22 Jul 2025 03:58:33 GMT"
},
{
"version": "v6",
"created": "Tue, 29 Jul 2025 03:20:01 GMT"
},
{
"version": "v7",
"created": "Thu, 07 Aug 2025 07:58:55 GMT"
},
{
"version": "v8",
"created": "Wed, 20 Aug 2025 10:11:43 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Banakh",
"Taras",
""
]
] |
2506.15059
|
Thor Tepper-Garc\'ia
|
Thor Tepper-Garc\'ia, Joss Bland-Hawthorn, Timothy R. Bedding, Christoph Federrath and Oscar Agertz
|
Galactic seismology: can the Gaia 'phase spiral' co-exist with a clumpy, turbulent interstellar medium?
|
MNRAS accepted (PDF matches accepted version)
| null | null | null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Gaia satellite revealed a remarkable spiral pattern ('phase spiral', PS) in the z-Vz phase-plane throughout the solar neighbourhood, where z and Vz are the displacement and velocity of a star perpendicular to the Galactic plane. As demonstrated by many groups, the kinematic signature reflects the Galactic stellar disc's response to a dynamical disturbance some 0.3-3 Gyr ago. However, previous controlled simulations did not consider the impact of the multi-phase interstellar medium (ISM) on the existence of the PS. This is crucial because it has been suggested that this weak signal is highly susceptible to scattering by small-scale density fluctuations typical of the ISM. This has motivated us to explore the formation and fate of the PS in a suite of high-resolution, N-body/hydrodynamical simulations of an idealised Galaxy analogue bearing a realistic ISM that interacts impulsively with a massive perturber. In our models, high gas surface densities within the disc encourage vigorous star formation, which in turn couples with the gas via feedback to drive turbulence. We find that the PS is almost non-existent if the disc potential is too strong or the ISM is highly structured on sub-kiloparsec scales. This can happen in the absence of stellar feedback when the gas is allowed to cool. In the presence of turbulent gas maintained by stellar feedback, the PS has a patchy spatial distribution and a high degree of intermittency on kiloparsec scales. We anticipate that future studies of the phase-spiral behaviour on all scales will provide crucial information on star-gas dynamics.
|
[
{
"version": "v1",
"created": "Wed, 18 Jun 2025 01:56:44 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:19:30 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tepper-García",
"Thor",
""
],
[
"Bland-Hawthorn",
"Joss",
""
],
[
"Bedding",
"Timothy R.",
""
],
[
"Federrath",
"Christoph",
""
],
[
"Agertz",
"Oscar",
""
]
] |
2506.15072
|
Zhenyuan Zhang
|
Jose Blanchet, Wei Cai, Shaswat Mohanty, Zhenyuan Zhang
|
Large Deviations of First Passage Times of Branching Random Walks in $\mathbb{R}^d$: Asymptotics and Algorithms
|
33 pages, 8 figures
| null | null | null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the large deviation probabilities of first passage times (FPT) of discrete-time supercritical non-lattice branching random walks (BRWs) in $\mathbb{R}^d$ where $d\geq 1$. The FPT refers to the first time the BRW enters a ball of radius one with a distance $x$ from the origin, conditioned upon the process's survival. Furthermore, we apply the spine decomposition technique to construct an asymptotically optimal polynomial-time algorithm for computing the lower large deviation probabilities of the FPT. The accuracy of our algorithm is also verified numerically. Our analysis not only provides a deeper theoretical understanding of these stochastic processes but also offers new insights into the microstructural features that are key to characterizing the strength of polymers.
|
[
{
"version": "v1",
"created": "Wed, 18 Jun 2025 02:29:18 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 22:14:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Blanchet",
"Jose",
""
],
[
"Cai",
"Wei",
""
],
[
"Mohanty",
"Shaswat",
""
],
[
"Zhang",
"Zhenyuan",
""
]
] |
2506.15900
|
Tyler Barna
|
Tyler Barna, Christoffer Fremling, Tomas Ahumada, Igor Andreoni, Smaranika Banerjee, Joshua S. Bloom, Mattia Bulla, Tracy X. Chen, Michael W. Coughlin, Tim Dietrich, Xander J. Hall, Alexandra Junell, Ben Rusholme, Jesper Sollerman, Niharika Sravan
|
IIb or not IIb: A Catalog of ZTF Kilonova Imposters
|
25 pages, 12 figures, 1 table, Submitted to PASP
| null |
10.1088/1538-3873/adf578
| null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Among the various classes of fast optical transients (FOTs), kilonovae (KNe), which can emerge as a result of neutron star mergers, are extremely challenging to observe because of not only the rapid timescale on which they fade (on the order of days), but also due to the relative scarcity of their occurrence. This scarcity is compounded by the large number of other FOTs that may initially resemble the characteristic rise of a KNe. While these objects can be ruled out as candidate KNe by taking spectroscopy, a method of confidently ruling out transients based on photometric analysis alone would be incredibly valuable. We describe the compilation of various "imposter" transients, including a plurality of IIb SNe, and investigate a number of comparative metrics by which one might be able to remove transients from consideration without the use of spectroscopy. We provide a list of these objects and their classifications as well as a glossary of the transient types included in the sample.
|
[
{
"version": "v1",
"created": "Wed, 18 Jun 2025 22:01:14 GMT"
},
{
"version": "v2",
"created": "Tue, 22 Jul 2025 16:06:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Barna",
"Tyler",
""
],
[
"Fremling",
"Christoffer",
""
],
[
"Ahumada",
"Tomas",
""
],
[
"Andreoni",
"Igor",
""
],
[
"Banerjee",
"Smaranika",
""
],
[
"Bloom",
"Joshua S.",
""
],
[
"Bulla",
"Mattia",
""
],
[
"Chen",
"Tracy X.",
""
],
[
"Coughlin",
"Michael W.",
""
],
[
"Dietrich",
"Tim",
""
],
[
"Hall",
"Xander J.",
""
],
[
"Junell",
"Alexandra",
""
],
[
"Rusholme",
"Ben",
""
],
[
"Sollerman",
"Jesper",
""
],
[
"Sravan",
"Niharika",
""
]
] |
2506.15930
|
Hantao Zhang
|
Hantao Zhang and Ran Cheng
|
Characterizing Spin-Orbit Torques by Tensorial Spin Hall Magnetoresistance
| null | null | null | null |
cond-mat.mes-hall cond-mat.mtrl-sci cond-mat.other
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Magnetoresistance (MR) provides a crucial tool for experimentally studying spin torques. While MR is well established in the device geometry of the spin Hall effect (SHE), as exemplified by the magnet/heavy-metal heterostructures, its role and manifestation beyond the SHE paradigm remain elusive. We propose a hitherto unknown form of MR where the underlying charge-to-spin conversion and its inverse process violate the simple geometry of the SHE, calling for tensorial descriptions. This MR can generate a series of unique harmonic responses essential for the experimental characterization of unconventional spin-orbit torques in non-SHE materials. We demonstrate these harmonic signals with semimetal WTe$_2$ in mind but the results are not restricted to specific materials.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 00:15:17 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 19:01:24 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhang",
"Hantao",
""
],
[
"Cheng",
"Ran",
""
]
] |
2506.16095
|
Rajan Das Gupta
|
Xun Liu, Xiaobin Wu, Jiaqi He, Rajan Das Gupta
|
Intelligent Operation and Maintenance and Prediction Model Optimization for Improving Wind Power Generation Efficiency
|
7 pages, 3 figures
|
Proc. 7th Int. Congr. on Human-Computer Interaction, Optimization and Robotic Applications (ICHORA), IEEE, pp. 1-7, 2025
|
10.1109/ICHORA65333.2025.11017307
| null |
eess.SY cs.SY stat.AP
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This study explores the effectiveness of predictive maintenance models and the optimization of intelligent Operation and Maintenance (O&M) systems in improving wind power generation efficiency. Through qualitative research, structured interviews were conducted with five wind farm engineers and maintenance managers, each with extensive experience in turbine operations. Using thematic analysis, the study revealed that while predictive maintenance models effectively reduce downtime by identifying major faults, they often struggle with detecting smaller, gradual failures. Key challenges identified include false positives, sensor malfunctions, and difficulties in integrating new models with older turbine systems. Advanced technologies such as digital twins, SCADA systems, and condition monitoring have significantly enhanced turbine maintenance practices. However, these technologies still require improvements, particularly in AI refinement and real-time data integration. The findings emphasize the need for continuous development to fully optimize wind turbine performance and support the broader adoption of renewable energy.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 07:31:03 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Liu",
"Xun",
""
],
[
"Wu",
"Xiaobin",
""
],
[
"He",
"Jiaqi",
""
],
[
"Gupta",
"Rajan Das",
""
]
] |
2506.16220
|
Nathan Szwarcberg
|
Nathan Szwarcberg (LMA), Tom Colinot, Christophe Vergez (LMA), Micha\"el Jousserand
|
Geometric sensitivity of modal parameters in wind instrument models: a case study on saxophone intonation
|
Acta Acustica
| null | null | null |
physics.class-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Transfer Matrix Method is a practical approach for modeling plane wave propagation in one-dimensional waveguides. Its simplicity makes it especially attractive for accounting for viscothermal losses, enabling realistic simulations of complex waveguides such as wind instruments. Another strength of this method lies in its fully analytical formulation of wave propagation. Modal parameters naturally arise as by-products of the model, obtained by numerically solving analytical expressions. In this work, the analytical potential of the method is extended by deriving the sensitivity of modal parameters to changes in the geometry of the resonator. These analytical gradients are applied in the context of wind instrument design. A simplified model of a soprano saxophone is used to investigate how octave harmonicity can be optimized through small geometric adjustments. The proposed approach enables predictive adjustments of geometry and offers valuable insight for both sound synthesis and instrument making.
|
[
{
"version": "v1",
"created": "Thu, 19 Jun 2025 11:24:23 GMT"
},
{
"version": "v2",
"created": "Wed, 02 Jul 2025 09:42:54 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 11:53:24 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Szwarcberg",
"Nathan",
"",
"LMA"
],
[
"Colinot",
"Tom",
"",
"LMA"
],
[
"Vergez",
"Christophe",
"",
"LMA"
],
[
"Jousserand",
"Michaël",
""
]
] |
2506.16812
|
Eduardo Ribas Brito
|
Dan Bogdanov, Eduardo Brito, Annika Jaakson, Peeter Laud and Raul-Martin Rebane
|
Zero-Knowledge Proof-of-Location Protocols for Vehicle Subsidies and Taxation Compliance
|
This is the extended version of the paper to appear in the Proceedings of the 5th International Workshop on Security and Privacy in Intelligent Infrastructures (SP2I 2025), held in conjunction with the 20th International Conference on Availability, Reliability and Security (ARES 2025)
| null |
10.1007/978-3-032-00642-4_20
| null |
cs.CR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper introduces a new set of privacy-preserving mechanisms for verifying compliance with location-based policies for vehicle taxation, or for (electric) vehicle (EV) subsidies, using Zero-Knowledge Proofs (ZKPs). We present the design and evaluation of a Zero-Knowledge Proof-of-Location (ZK-PoL) system that ensures a vehicle's adherence to territorial driving requirements without disclosing specific location data, hence maintaining user privacy. Our findings suggest a promising approach to apply ZK-PoL protocols in large-scale governmental subsidy or taxation programs.
|
[
{
"version": "v1",
"created": "Fri, 20 Jun 2025 08:00:32 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bogdanov",
"Dan",
""
],
[
"Brito",
"Eduardo",
""
],
[
"Jaakson",
"Annika",
""
],
[
"Laud",
"Peeter",
""
],
[
"Rebane",
"Raul-Martin",
""
]
] |
2506.17413
|
Sourish Banerjee
|
Sourish Banerjee, Rudranil Basu
|
Light and Shadow OPEs: A Carroll Symmetric Approach to Flat Holography
|
26 pages, typos corrected
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we fix the leading term in the operator algebra of light-transformed operators in the context of flat space holography. Starting with light-transformed graviton correlators, we show that the OPE obtained by taking the collinear limit satisfies translation symmetry at the leading order not independently but with assistance from the sub-leading order. Motivated by this result, we start with a general CFT-like ansatz for the OPE and, by keeping track of the sub-leading term, derive an expression for the scaling dimension of the operator appearing in the leading term. We further show that we can fix the OPE coefficient corresponding to this term as well. We use this result to find the OPE for well-studied theories like gravity, Yang-Mills and Einstein-Yang-Mills theory, and find that the scaling dimension found from our formula matches with the one obtained by looking at the collinear limit of bulk momentum space vertices. We initiate a similar study for operators in the shadow basis by looking at the OPE of shadow-transformed graviton correlators.
|
[
{
"version": "v1",
"created": "Fri, 20 Jun 2025 18:18:52 GMT"
},
{
"version": "v2",
"created": "Tue, 24 Jun 2025 19:45:18 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 07:32:57 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Banerjee",
"Sourish",
""
],
[
"Basu",
"Rudranil",
""
]
] |
2506.18897
|
Xiaowei Chi
|
Xiaowei Chi, Kuangzhi Ge, Jiaming Liu, Siyuan Zhou, Peidong Jia, Zichen He, Yuzhen Liu, Tingguang Li, Lei Han, Sirui Han, Shanghang Zhang, Yike Guo
|
MinD: Learning A Dual-System World Model for Real-Time Planning and Implicit Risk Analysis
| null | null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Video Generation Models (VGMs) have become powerful backbones for Vision-Language-Action (VLA) models, leveraging large-scale pretraining for robust dynamics modeling. However, current methods underutilize their distribution modeling capabilities for predicting future states. Two challenges hinder progress: integrating generative processes into feature learning is both technically and conceptually underdeveloped, and naive frame-by-frame video diffusion is computationally inefficient for real-time robotics. To address these, we propose Manipulate in Dream (MinD), a dual-system world model for real-time, risk-aware planning. MinD uses two asynchronous diffusion processes: a low-frequency visual generator (LoDiff) that predicts future scenes and a high-frequency diffusion policy (HiDiff) that outputs actions. Our key insight is that robotic policies do not require fully denoised frames but can rely on low-resolution latents generated in a single denoising step. To connect early predictions to actions, we introduce DiffMatcher, a video-action alignment module with a novel co-training strategy that synchronizes the two diffusion models. MinD achieves a 63% success rate on RL-Bench, 60% on real-world Franka tasks, and operates at 11.3 FPS, demonstrating the efficiency of single-step latent features for control signals. Furthermore, MinD identifies 74% of potential task failures in advance, providing real-time safety signals for monitoring and intervention. This work establishes a new paradigm for efficient and reliable robotic manipulation using generative world models.
|
[
{
"version": "v1",
"created": "Mon, 23 Jun 2025 17:59:06 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:07:13 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chi",
"Xiaowei",
""
],
[
"Ge",
"Kuangzhi",
""
],
[
"Liu",
"Jiaming",
""
],
[
"Zhou",
"Siyuan",
""
],
[
"Jia",
"Peidong",
""
],
[
"He",
"Zichen",
""
],
[
"Liu",
"Yuzhen",
""
],
[
"Li",
"Tingguang",
""
],
[
"Han",
"Lei",
""
],
[
"Han",
"Sirui",
""
],
[
"Zhang",
"Shanghang",
""
],
[
"Guo",
"Yike",
""
]
] |
2506.19443
|
Ayush Kumar Tewari
|
Jian-Rong Li and Ayush Kumar Tewari
|
From dual canonical bases to positroidal subdivisions
| null | null | null | null |
math.CO math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
The Grassmannian cluster algebra $\mathbb{C}[\text{Gr}(k, n)]$ admits a distinguished basis known as the dual canonical basis, whose elements correspond to rectangular semi-standard Young tableaux with $k$ rows and with entries in $[n]$. We establish that each such tableau induces a positroidal subdivision of the hypersimplex $\Delta(k,n)$ via a map introduced by Speyer and Williams. For $\text{Gr}(2,n)$, we prove that non-frozen prime tableaux correspond precisely to the coarsest positroidal subdivisions of $\Delta(2,n)$. Furthermore, we present computational evidence extending these results to $k>2$. In the process, we formulate a conjectural formula for the number of split positroidal subdivisions of $\Delta(k,n)$ for any $k \ge 2$ and explore the deep connections between the polyhedral combinatorics of $\Delta(k,n)$ and the dual canonical basis of $\mathbb{C}[\text{Gr}(k, n)]$.
|
[
{
"version": "v1",
"created": "Tue, 24 Jun 2025 09:16:03 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:55:59 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Li",
"Jian-Rong",
""
],
[
"Tewari",
"Ayush Kumar",
""
]
] |
2506.20380
|
Madeline Lisaius
|
Zhengpeng Feng, Clement Atzberger, Sadiq Jaffer, Jovana Knezevic, Silja Sormunen, Robin Young, Madeline C Lisaius, Markus Immitzer, James Ball, David A. Coomes, Anil Madhavapeddy, Andrew Blake and Srinivasan Keshav
|
TESSERA: Temporal Embeddings of Surface Spectra for Earth Representation and Analysis
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Satellite remote sensing enables a wide range of downstream applications, including habitat mapping, carbon accounting, and strategies for conservation and sustainable land use. However, satellite time series are voluminous and often corrupted, making them challenging to use. We present TESSERA, an open, global, land-oriented remote sensing foundation model that uses self-supervised learning to generate `ready-to-use' embeddings at 10~m scale from pixel-level satellite time-series data. TESSERA uses two encoders to combine optical data with synthetic aperture radar backscatter coefficients at 10~m resolution to create embeddings that are fused with a multilayer perceptron to create annual global embedding maps. We compare our work with state-of-the-art task-specific models and other foundation models in five diverse downstream tasks and find that TESSERA closely matches or outperforms these baselines. We believe that TESSERA's ease of use, state-of-the-art performance, openness, and computation- and labelled data-efficiency will prove transformative in a wide range of ecological applications.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 12:46:26 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Jul 2025 17:22:48 GMT"
},
{
"version": "v3",
"created": "Tue, 29 Jul 2025 15:23:52 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 16:28:55 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Feng",
"Zhengpeng",
""
],
[
"Atzberger",
"Clement",
""
],
[
"Jaffer",
"Sadiq",
""
],
[
"Knezevic",
"Jovana",
""
],
[
"Sormunen",
"Silja",
""
],
[
"Young",
"Robin",
""
],
[
"Lisaius",
"Madeline C",
""
],
[
"Immitzer",
"Markus",
""
],
[
"Ball",
"James",
""
],
[
"Coomes",
"David A.",
""
],
[
"Madhavapeddy",
"Anil",
""
],
[
"Blake",
"Andrew",
""
],
[
"Keshav",
"Srinivasan",
""
]
] |
2506.20857
|
James Beck
|
James L Beck
|
Free Electron Paths from Dirac's Wave Equation Elucidating Zitterbewegung and Spin
|
26 pages. No figures
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The worldline of a free electron is revealed by applying Dirac's velocity operator to its Dirac wave function whose space-time arguments are expressed in a proper time by a Lorentz transformation. This motion can be decomposed into two parts: the electron's global motion of its inertia (or spin) center and an inherent local periodic motion about this point that produces the electron's spin and has the zitterbewegung frequency found by Schr\"{o}dinger in his operator analysis of Dirac's wave equation. This zitter motion corresponds to the so-called polarization and magnetization currents in Gordon's decomposition of Dirac's current. In an inertial "rest"-frame fixed at the inertia center, Dirac's wave function for a free electron with its spin in a specified direction implies that the zitter motion is a perpetual circular motion about the inertia center in a plane orthogonal to this spin direction with a radius one half of the Compton radius and moving at the speed of light. The electron continuously accelerates about the spin center without any external force because the inertia is effective at the spin center, rather than at its charge center where the electron interacts with the electro-magnetic field. This analysis confirms the nature of zitterbewegung directly from Dirac's wave equation, agreeing with the conclusions of Barut and Zanghi, Beck, Hestenes, Rivas and Salesi from their classical Dirac particle models of the electron. Furthermore, these five classical models are equivalent and express the same free electron dynamics as Dirac's equation.
|
[
{
"version": "v1",
"created": "Wed, 25 Jun 2025 22:09:08 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 22:42:33 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Beck",
"James L",
""
]
] |
2506.21194
|
Nataliia Kopteva
|
Carolin Harkort, Ina V. Kalitukha, Nataliia E. Kopteva, Mikhail O. Nestoklon, Serguei V. Goupalov, Lucien Saviot, Dennis Kudlacik, Dmitri R. Yakovlev, Elena V. Kolobkova, Maria S. Kuznetsova, and Manfred Bayer
|
Confined acoustic phonons in CsPbI3 nanocrystals explored by resonant Raman scattering on excitons
| null |
Nano Letters 25,12754 (2025)
|
10.1021/acs.nanolett.5c03342
| null |
cond-mat.mtrl-sci cond-mat.other
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Optical properties of the lead halide perovskites nanocrystals are controlled by confined excitons and rich spectrum of confined acoustic and optical phonons. We study experimentally and theoretically the exciton-phonon interaction in CsPbI3 perovskite nanocrystals embedded in a glass matrix. Energies of phonon modes allowed by selection rules are detected by resonant Raman scattering for nanocrystals with sizes of 4-13 nm, covering exciton energies of 1.72-2.25 eV. While optical phonon energies remain size-independent, the energies of confined acoustic phonons increase in smaller nanocrystals. Acoustic phonons are modeled within the continuum approximation using elastic constants computed by density functional theory. The model provides the energy spectra of confined phonons for nanocrystals of various shapes (cube, sphere, spheroid), crystal symmetries (orthorhombic and tetragonal), and sizes. Exciton confinement restricts efficient coupling to only few phonon modes observable in Raman spectra. By comparing experimental data with model predictions, we conclude that the nanocrystals in our samples predominantly have spherical or spheroidal shapes.
|
[
{
"version": "v1",
"created": "Thu, 26 Jun 2025 12:54:21 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Harkort",
"Carolin",
""
],
[
"Kalitukha",
"Ina V.",
""
],
[
"Kopteva",
"Nataliia E.",
""
],
[
"Nestoklon",
"Mikhail O.",
""
],
[
"Goupalov",
"Serguei V.",
""
],
[
"Saviot",
"Lucien",
""
],
[
"Kudlacik",
"Dennis",
""
],
[
"Yakovlev",
"Dmitri R.",
""
],
[
"Kolobkova",
"Elena V.",
""
],
[
"Kuznetsova",
"Maria S.",
""
],
[
"Bayer",
"Manfred",
""
]
] |
2506.21205
|
Elia Trevisan
|
Elia Trevisan, Khaled A. Mustafa, Godert Notten, Xinwei Wang and Javier Alonso-Mora
|
Dynamic Risk-Aware MPPI for Mobile Robots in Crowds via Efficient Monte Carlo Approximations
|
Accepted for presentation at IROS 2025. Accepted Version
| null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Deploying mobile robots safely among humans requires the motion planner to account for the uncertainty in the other agents' predicted trajectories. This remains challenging in traditional approaches, especially with arbitrarily shaped predictions and real-time constraints. To address these challenges, we propose a Dynamic Risk-Aware Model Predictive Path Integral control (DRA-MPPI), a motion planner that incorporates uncertain future motions modelled with potentially non-Gaussian stochastic predictions. By leveraging MPPI's gradient-free nature, we propose a method that efficiently approximates the joint Collision Probability (CP) among multiple dynamic obstacles for several hundred sampled trajectories in real-time via a Monte Carlo (MC) approach. This enables the rejection of samples exceeding a predefined CP threshold or the integration of CP as a weighted objective within the navigation cost function. Consequently, DRA-MPPI mitigates the freezing robot problem while enhancing safety. Real-world and simulated experiments with multiple dynamic obstacles demonstrate DRA-MPPI's superior performance compared to state-of-the-art approaches, including Scenario-based Model Predictive Control (S-MPC), Frenet planner, and vanilla MPPI.
|
[
{
"version": "v1",
"created": "Thu, 26 Jun 2025 13:01:32 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:38:26 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Trevisan",
"Elia",
""
],
[
"Mustafa",
"Khaled A.",
""
],
[
"Notten",
"Godert",
""
],
[
"Wang",
"Xinwei",
""
],
[
"Alonso-Mora",
"Javier",
""
]
] |
2506.21265
|
Elia Trevisan
|
Jelmer van der Saag, Elia Trevisan, Wouter Falkena and Javier Alonso-Mora
|
Active Disturbance Rejection Control for Trajectory Tracking of a Seagoing USV: Design, Simulation, and Field Experiments
|
Accepted for presentation at IROS 2025. Accepted version
| null | null | null |
cs.RO cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Unmanned Surface Vessels (USVs) face significant control challenges due to uncertain environmental disturbances like waves and currents. This paper proposes a trajectory tracking controller based on Active Disturbance Rejection Control (ADRC) implemented on the DUS V2500. A custom simulation incorporating realistic waves and current disturbances is developed to validate the controller's performance, supported by further validation through field tests in the harbour of Scheveningen, the Netherlands, and at sea. Simulation results demonstrate that ADRC significantly reduces cross-track error across all tested conditions compared to a baseline PID controller but increases control effort and energy consumption. Field trials confirm these findings while revealing a further increase in energy consumption during sea trials compared to the baseline.
|
[
{
"version": "v1",
"created": "Thu, 26 Jun 2025 13:48:13 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:44:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"van der Saag",
"Jelmer",
""
],
[
"Trevisan",
"Elia",
""
],
[
"Falkena",
"Wouter",
""
],
[
"Alonso-Mora",
"Javier",
""
]
] |
2506.22562
|
Abhineet Singh
|
Abhineet Singh and Nilanjan Ray
|
Improving Token-based Object Detection with Video
|
Published in IEEE Access
| null |
10.1109/ACCESS.2025.3596215
| null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
This paper improves upon the Pix2Seq object detector by extending it for videos. In the process, it introduces a new way to perform end-to-end video object detection that improves upon existing video detectors in two key ways. First, by representing objects as variable-length sequences of discrete tokens, we can succinctly represent widely varying numbers of video objects, with diverse shapes and locations, without having to inject any localization cues in the training process. This eliminates the need to sample the space of all possible boxes that constrains conventional detectors and thus solves the dual problems of loss sparsity during training and heuristics-based postprocessing during inference. Second, it conceptualizes and outputs the video objects as fully integrated and indivisible 3D boxes or tracklets instead of generating image-specific 2D boxes and linking these boxes together to construct the video object, as done in most conventional detectors. This allows it to scale effortlessly with available computational resources by simply increasing the length of the video subsequence that the network takes as input, even generalizing to multi-object tracking if the subsequence can span the entire video. We compare our video detector with the baseline Pix2Seq static detector on several datasets and demonstrate consistent improvement, although with strong signs of being bottlenecked by our limited computational resources. We also compare it with several video detectors on UA-DETRAC to show that it is competitive with the current state of the art even with the computational bottleneck. We make our code and models publicly available.
|
[
{
"version": "v1",
"created": "Fri, 27 Jun 2025 18:21:58 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:50:02 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Singh",
"Abhineet",
""
],
[
"Ray",
"Nilanjan",
""
]
] |
2506.22734
|
Gary Mart\'inez-N\'unez
|
Gary Martinez-Nunez
|
Proper polyhedral divisors and torus actions over arbitrary fields
|
69 pages, comments are welcome :)
| null | null | null |
math.AG
|
http://creativecommons.org/publicdomain/zero/1.0/
|
We provide a algebro-geometric combinatorial description of geometrically integral geometrically normal affine varieties endowed with an effective action of an algebraic torus over arbitrary fields. This description is achieved in terms of proper polyhedral divisors endowed with a Galois semilinear action.
|
[
{
"version": "v1",
"created": "Sat, 28 Jun 2025 02:42:20 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 15:57:28 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Martinez-Nunez",
"Gary",
""
]
] |
2506.23036
|
Zain Ul Abdeen
|
Zain ul Abdeen, Ming Jin
|
Fragile, Robust, and Antifragile: A Perspective from Parameter Responses in Reinforcement Learning Under Stress
|
Withdrawn pending a review of attribution and overlap with Pravin et al., Artificial Intelligence (2024), DOI: 10.1016/j.artint.2023.104060. Further dissemination is paused while we determine appropriate next steps
| null | null | null |
cs.LG cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper explores Reinforcement learning (RL) policy robustness by systematically analyzing network parameters under internal and external stresses. Inspired by synaptic plasticity in neuroscience, synaptic filtering introduces internal stress by selectively perturbing parameters, while adversarial attacks apply external stress through modified agent observations. This dual approach enables the classification of parameters as fragile, robust, or antifragile, based on their influence on policy performance in clean and adversarial settings. Parameter scores are defined to quantify these characteristics, and the framework is validated on PPO-trained agents in Mujoco continuous control environments. The results highlight the presence of antifragile parameters that enhance policy performance under stress, demonstrating the potential of targeted filtering techniques to improve RL policy adaptability. These insights provide a foundation for future advancements in the design of robust and antifragile RL systems.
|
[
{
"version": "v1",
"created": "Sat, 28 Jun 2025 23:28:47 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:21:01 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Abdeen",
"Zain ul",
""
],
[
"Jin",
"Ming",
""
]
] |
2507.00497
|
Takami Tohyama
|
Takami Tohyama, Hidetoshi Fukuyama
|
Effects of Antisite Defects on Seebeck Coefficient in Fe_2VAl -- Analyses based on Bipolar Random Anderson Model
|
7 pages, 8 figures, to be published in J. Phys. Soc. Jpn. vol. 94, (2025)
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
A microscopic mechanism is proposed for a dramatic sign change of the Seebeck coefficient from positive to negative sign by the introduction of antisite defects in Fe$_2$VAl based on bipolar random Anderson model (BPRAM), which incorporates hybridization effects between randomly distributed antisites and host bands, where the valence and conduction bands are treated separately due to their separation in momentum space. Applying a self-consistent T-matrix approximation, we find that antisite defects in Fe$_2$VAl induce new states in the band overlap region, resulting in a scattering rate that is higher for hole carriers in the valence band than that for electron carriers in the conduction band, leading to negative Seebeck coefficient. This mechanism of sign change presents a potential new approach for controlling thermoelectric properties in semimetallic systems without changing carrier concentration.
|
[
{
"version": "v1",
"created": "Tue, 01 Jul 2025 07:13:13 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:41:38 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tohyama",
"Takami",
""
],
[
"Fukuyama",
"Hidetoshi",
""
]
] |
2507.02251
|
Fritz Gesztesy
|
Fritz Gesztesy and Roger Nichols
|
A generalized Birman-Schwinger principle and applications to one-dimensional Schr\"odinger operators with distributional potentials
|
29 pages
| null | null | null |
math.FA math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
Given a self-adjoint operator $H_0$ bounded from below in a complex Hilbert space $\mathcal{H}$, the corresponding scale of spaces $\mathcal{H}_{+1}(H_0) \subset \mathcal{H} \subset \mathcal{H}_{-1}(H_0) = [\mathcal{H}_{+1}(H_0)]^*$, and a fixed $V\in \mathcal{B}(\mathcal{H}_{+1}(H_0),\mathcal{H}_{-1}(H_0))$, we define the operator-valued map $A_V(\,\cdot\,):\rho(H_0)\to \mathcal{B}(\mathcal{H})$ by \[ A_V(z):=-\big(H_0-zI_{\mathcal{H}} \big)^{-1/2}V\big(H_0-zI_{\mathcal{H}} \big)^{-1/2}\in \mathcal{B}(\mathcal{H}),\quad z\in \rho(H_0), \] where $\rho(H_0)$ denotes the resolvent set of $H_0$. Assuming that $A_V(z)$ is compact for some $z=z_0\in \rho(H_0)$ and has norm strictly less than one for some $z=E_0\in (-\infty,0)$, we employ an abstract version of Tiktopoulos' formula to define an operator $H$ in $\mathcal{H}$ that is formally realized as the sum of $H_0$ and $V$. We then establish a Birman-Schwinger principle for $H$ in which $A_V(\,\cdot\,)$ plays the role of the Birman-Schwinger operator: $\lambda_0\in \rho(H_0)$ is an eigenvalue of $H$ if and only if $1$ is an eigenvalue of $A_V(\lambda_0)$. Furthermore, the geometric (but not necessarily the algebraic) multiplicities of $\lambda_0$ and $1$ as eigenvalues of $H$ and $A_V(\lambda_0)$, respectively, coincide.
As a concrete application, we consider one-dimensional Schr\"odinger operators with $H^{-1}(\mathbb{R})$ distributional potentials.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 02:59:16 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:50:59 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gesztesy",
"Fritz",
""
],
[
"Nichols",
"Roger",
""
]
] |
2507.02755
|
Caleb Rascon
|
Caleb Rascon, Luis Gato-Diaz, Eduardo Garc\'ia-Alarc\'on
|
Multi-agent Auditory Scene Analysis
|
Submitted to Applied Soft Computing
| null | null | null |
eess.AS cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Auditory scene analysis (ASA) aims to retrieve information from the acoustic environment, by carrying out three main tasks: sound source location, separation, and classification. These tasks are traditionally executed with a linear data flow, where the sound sources are first located; then, using their location, each source is separated into its own audio stream; from each of which, information is extracted that is relevant to the application scenario (audio event detection, speaker identification, emotion classification, etc.). However, running these tasks linearly increases the overall response time, while making the last tasks (separation and classification) highly sensitive to errors of the first task (location). A considerable amount of effort and computational complexity has been employed in the state-of-the-art to develop techniques that are the least error-prone possible. However, doing so gives rise to an ASA system that is non-viable in many applications that require a small computational footprint and a low response time, such as bioacoustics, hearing-aid design, search and rescue, human-robot interaction, etc. To this effect, in this work, a multi-agent approach is proposed to carry out ASA where the tasks are run in parallel, with feedback loops between them to compensate for local errors, such as: using the quality of the separation output to correct the location error; and using the classification result to reduce the localization's sensitivity towards interferences. The result is a multi-agent auditory scene analysis (MASA) system that is robust against local errors, without a considerable increase in complexity, and with a low response time. The complete proposed MASA system is provided as a publicly available framework that uses open-source tools for sound acquisition and reproduction (JACK) and inter-agent communication (ROS2), allowing users to add their own agents.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 16:16:46 GMT"
},
{
"version": "v2",
"created": "Sun, 17 Aug 2025 19:24:34 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 14:18:03 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Rascon",
"Caleb",
""
],
[
"Gato-Diaz",
"Luis",
""
],
[
"García-Alarcón",
"Eduardo",
""
]
] |
2507.03047
|
Yutian Liu
|
Yutian Liu, Zhengyi Yang, Jiancan Wu, Xiang Wang
|
Enhancing Temporal Sensitivity of Large Language Model for Recommendation with Counterfactual Tuning
| null | null | null | null |
cs.CL cs.AI cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances have applied large language models (LLMs) to sequential recommendation, leveraging their pre-training knowledge and reasoning capabilities to provide more personalized user experiences. However, existing LLM-based methods fail to sufficiently leverage the rich temporal information inherent in users' historical interaction sequences, stemming from fundamental architectural constraints: LLMs process information through self-attention mechanisms that lack inherent sequence ordering and rely on position embeddings designed primarily for natural language rather than user interaction sequences. This limitation significantly impairs their ability to capture the evolution of user preferences over time and predict future interests accurately.
To address this critical gap, we propose \underline{C}ounterfactual \underline{E}nhanced \underline{T}emporal Framework for LLM-Based \underline{Rec}ommendation (CETRec). CETRec is grounded in causal inference principles, which allow it to isolate and measure the specific impact of temporal information on recommendation outcomes. Combined with our counterfactual tuning task derived from causal analysis, CETRec effectively enhances LLMs' awareness of both absolute order (how recently items were interacted with) and relative order (the sequential relationships between items). Extensive experiments on real-world datasets demonstrate the effectiveness of our CETRec. Our code is available at https://anonymous.4open.science/r/CETRec-B9CE/.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 10:11:35 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 09:09:56 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Liu",
"Yutian",
""
],
[
"Yang",
"Zhengyi",
""
],
[
"Wu",
"Jiancan",
""
],
[
"Wang",
"Xiang",
""
]
] |
2507.03090
|
Georges Obied
|
Alek Bedroya, Georges Obied, Cumrun Vafa, David H. Wu
|
Evolving Dark Sector and the Dark Dimension Scenario
|
13 pages, 7 figures; v2: minor edits, added refs, added supplementary materials studying simple variations of the original model (22 pages, 15 figures)
| null | null | null |
astro-ph.CO gr-qc hep-ph hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
String theory naturally leads to the expectation that dark energy is not stable, and may be evolving as captured by the Swampland de Sitter conjectures. Moreover, motivated by the distance conjecture, a unification of dark sector has been proposed, where the smallness of dark energy leads to one extra dimension of micron size with dark matter being the Kaluza--Klein graviton excitations in this extra dimension. We consider the natural possibility that the radius of the dark dimension varies as the dark energy decreases, leading to the variation of the dark matter mass. This correlates the decrease of the dark energy with the variation of the dark matter mass as they depend on the variations of a scalar field $\phi$ controlling the radius of the extra dimension. A simple realization of this idea for small range of $\phi$ is adequately captured by choosing a potential which is locally of the form $V=V_0\ {\rm exp}(-c\phi)$ and dark matter mass $m_{\rm DM}=m_0\ {\rm exp}(-c' \phi)$ where the sign of $\phi$ is chosen such that $c'\geq 0$ while we have two choices for the sign of $c$ depending on whether the dark dimension expands or shrinks when the dark energy dominates. We find excellent agreement with recent experimental data from DESI DR2 combined with SN measurements (from DES, Union3 or Pantheon+) and reproduces the same significance as CPL parametrization with the added benefit of providing a natural explanation for the apparent phantom behavior ($w<-1$) reported by DESI and DES based on a physical model. DESI and SN datasets independently favor non-zero values of $c'$ and $c$, respectively, both lying within the expected $\mathcal{O}(1)$ range suggested by the Swampland criteria. Moreover, our best fit value $c'\simeq 0.05 \pm 0.01$ is remarkably consistent with the experimental upper bound of $c'\lesssim 0.2$ demanded by the lack of detection of fifth force in the dark sector.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 18:03:20 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 14:59:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bedroya",
"Alek",
""
],
[
"Obied",
"Georges",
""
],
[
"Vafa",
"Cumrun",
""
],
[
"Wu",
"David H.",
""
]
] |
2507.03608
|
Sarat Ahmad
|
Sarat Ahmad, Zeinab Nezami, Maryam Hafeez, Syed Ali Raza Zaidi
|
Benchmarking Vector, Graph and Hybrid Retrieval Augmented Generation (RAG) Pipelines for Open Radio Access Networks (ORAN)
| null | null | null | null |
cs.AI cs.DC cs.ET cs.NI
|
http://creativecommons.org/licenses/by/4.0/
|
Generative AI (GenAI) is expected to play a pivotal role in enabling autonomous optimization in future wireless networks. Within the ORAN architecture, Large Language Models (LLMs) can be specialized to generate xApps and rApps by leveraging specifications and API definitions from the RAN Intelligent Controller (RIC) platform. However, fine-tuning base LLMs for telecom-specific tasks remains expensive and resource-intensive. Retrieval-Augmented Generation (RAG) offers a practical alternative through in-context learning, enabling domain adaptation without full retraining. While traditional RAG systems rely on vector-based retrieval, emerging variants such as GraphRAG and Hybrid GraphRAG incorporate knowledge graphs or dual retrieval strategies to support multi-hop reasoning and improve factual grounding. Despite their promise, these methods lack systematic, metric-driven evaluations, particularly in high-stakes domains such as ORAN. In this study, we conduct a comparative evaluation of Vector RAG, GraphRAG, and Hybrid GraphRAG using ORAN specifications. We assess performance across varying question complexities using established generation metrics: faithfulness, answer relevance, context relevance, and factual correctness. Results show that both GraphRAG and Hybrid GraphRAG outperform traditional RAG. Hybrid GraphRAG improves factual correctness by 8%, while GraphRAG improves context relevance by 11%.
|
[
{
"version": "v1",
"created": "Fri, 04 Jul 2025 14:31:30 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:37:28 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ahmad",
"Sarat",
""
],
[
"Nezami",
"Zeinab",
""
],
[
"Hafeez",
"Maryam",
""
],
[
"Zaidi",
"Syed Ali Raza",
""
]
] |
2507.03984
|
Jeonghyo Song
|
Jeonghyo Song, Kimin Yun, DaeUng Jo, Jinyoung Kim, Youngjoon Yoo
|
CoT-Segmenter: Enhancing OOD Detection in Dense Road Scenes via Chain-of-Thought Reasoning
|
6 pages, 3 figures. Accepted at IEEE International Conference on Advanced Visual and Signal-Based Systems 2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Effective Out-of-Distribution (OOD) detection is criti-cal for ensuring the reliability of semantic segmentation models, particularly in complex road environments where safety and accuracy are paramount. Despite recent advancements in large language models (LLMs), notably GPT-4, which significantly enhanced multimodal reasoning through Chain-of-Thought (CoT) prompting, the application of CoT-based visual reasoning for OOD semantic segmentation remains largely unexplored. In this paper, through extensive analyses of the road scene anomalies, we identify three challenging scenarios where current state-of-the-art OOD segmentation methods consistently struggle: (1) densely packed and overlapping objects, (2) distant scenes with small objects, and (3) large foreground-dominant objects. To address the presented challenges, we propose a novel CoT-based framework targeting OOD detection in road anomaly scenes. Our method leverages the extensive knowledge and reasoning capabilities of foundation models, such as GPT-4, to enhance OOD detection through improved image understanding and prompt-based reasoning aligned with observed problematic scene attributes. Extensive experiments show that our framework consistently outperforms state-of-the-art methods on both standard benchmarks and our newly defined challenging subset of the RoadAnomaly dataset, offering a robust and interpretable solution for OOD semantic segmentation in complex driving environments.
|
[
{
"version": "v1",
"created": "Sat, 05 Jul 2025 10:23:40 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 10:14:24 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Song",
"Jeonghyo",
""
],
[
"Yun",
"Kimin",
""
],
[
"Jo",
"DaeUng",
""
],
[
"Kim",
"Jinyoung",
""
],
[
"Yoo",
"Youngjoon",
""
]
] |
2507.04164
|
Yimeng Min
|
Yimeng Min, Carla P. Gomes
|
Structure As Search: Unsupervised Permutation Learning for Combinatorial Optimization
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
We propose a non-autoregressive framework for the Travelling Salesman Problem where solutions emerge directly from learned permutations, without requiring explicit search. By applying a similarity transformation to Hamiltonian cycles, the model learns to approximate permutation matrices via continuous relaxations. Our unsupervised approach achieves competitive performance against classical heuristics, demonstrating that the inherent structure of the problem can effectively guide combinatorial optimization without sequential decision-making.
|
[
{
"version": "v1",
"created": "Sat, 05 Jul 2025 21:17:38 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 02:25:21 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Min",
"Yimeng",
""
],
[
"Gomes",
"Carla P.",
""
]
] |
2507.04487
|
Xujia Wang
|
Xujia Wang, Yunjia Qi, Bin Xu
|
LoSiA: Efficient High-Rank Fine-Tuning via Subnet Localization and Optimization
|
Accepted to EMNLP 2025 (Main Conference); 18 pages, 12 figures
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Parameter-Efficient Fine-Tuning (PEFT) methods, such as LoRA, significantly reduce the number of trainable parameters by introducing low-rank decomposition matrices. However, existing methods perform extensive matrix multiplications in domain specialization tasks, resulting in computational inefficiency and sub-optimal fine-tuning performance. Hence, we propose LoSiA(Low-Resources Subnet Integration Adaptation), an innovative method that dynamically localizes and optimizes critical parameters during the training process. Specifically, it identifies a sub-network using gradient sparsity analysis and optimizes it as the trainable target. This design enables effective high-rank adaptation by updating only the sub-network parameters, reducing the additional matrix multiplication. We also present LoSiA-Pro, a faster implementation of LoSiA, which reduces the training latency by about $27\%$ compared to LoRA. Extensive evaluations show that our method achieves minimal performance drop compared to full fine-tuning, while requiring the least training time across domain specialization and common-sense reasoning tasks. Further analysis shows that LoSiA also reduces forgetting during continued training. The source code is available at https://github.com/KlozeWang/LoSiA.
|
[
{
"version": "v1",
"created": "Sun, 06 Jul 2025 17:51:57 GMT"
},
{
"version": "v2",
"created": "Tue, 08 Jul 2025 05:22:57 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 17:33:10 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wang",
"Xujia",
""
],
[
"Qi",
"Yunjia",
""
],
[
"Xu",
"Bin",
""
]
] |
2507.04850
|
Tim Gershon
|
Tim Gershon, Thomas Latham, Peilian Li, Andy Morris, Wenbin Qian, Mark Whitehead, Ao Xu
|
More visualisation of decay-time-dependent asymmetries in multibody B-meson decays
|
Extension of method presented in arXiv:2401.13473 by same authors. 14 pages, 3 figures v2 minor updates including extra subfigure
| null | null | null |
hep-ph hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Methods have been proposed recently to weight data in order to allow visualisation of CP-violation effects in transitions of neutral B mesons to multibody final states that are not CP-eigenstates. These are useful since integration of the unweighted data over the phase space would otherwise wash out the effects of interest. A similar method, elaborated upon here, with a different weighting function can also be used to visualise CP-conserving $B_{(s)}^0$--$\overline{B}{}_{(s)}^0$ oscillations, rather than the CP-violating asymmetries. Together with the other weighting functions, this new method could be useful, for example, to demonstrate the accuracy of the calibration of the flavour tagging algorithms that are crucial for analyses such as the measurement of the CP-violating phase in $B_s^0 \to J/\!\psi\phi$ decays. Their application to the formalism in common use for such decays is explicitly demonstrated.
|
[
{
"version": "v1",
"created": "Mon, 07 Jul 2025 10:18:14 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 18:30:47 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gershon",
"Tim",
""
],
[
"Latham",
"Thomas",
""
],
[
"Li",
"Peilian",
""
],
[
"Morris",
"Andy",
""
],
[
"Qian",
"Wenbin",
""
],
[
"Whitehead",
"Mark",
""
],
[
"Xu",
"Ao",
""
]
] |
2507.04996
|
Jiangbo Yu
|
Jiangbo Yu
|
From Autonomy to Agency: Agentic Vehicles for Human-Centered Mobility Systems
| null | null | null | null |
cs.CY cs.CE cs.CL cs.HC cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Autonomy, from the Greek autos (self) and nomos (law), refers to the capacity to operate according to internal rules without external control. Accordingly, autonomous vehicles (AuVs) are viewed as vehicular systems capable of perceiving their environment and executing pre-programmed tasks independently of external input. However, both research and real-world deployments increasingly showcase vehicles that demonstrate behaviors beyond this definition (including the SAE levels 0 to 5); Examples of this outpace include the interaction with humans with natural language, goal adaptation, contextual reasoning, external tool use, and unseen ethical dilemma handling, largely empowered by multi-modal large language models (LLMs). These developments reveal a conceptual gap between technical autonomy and the broader cognitive and social capabilities needed for future human-centered mobility systems. To address this gap, this paper introduces the concept of agentic vehicles (AgVs), referring to vehicles that integrate agentic AI systems to reason, adapt, and interact within complex environments. This paper proposes the term AgVs and their distinguishing characteristics from conventional AuVs. It synthesizes relevant advances in integrating LLMs and AuVs and highlights how AgVs might transform future mobility systems and ensure the systems are human-centered. The paper concludes by identifying key challenges in the development and governance of AgVs, and how they can play a significant role in future agentic transportation systems.
|
[
{
"version": "v1",
"created": "Mon, 07 Jul 2025 13:34:49 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Aug 2025 19:04:36 GMT"
},
{
"version": "v3",
"created": "Fri, 15 Aug 2025 14:21:57 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 13:50:10 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Yu",
"Jiangbo",
""
]
] |
2507.06358
|
Victor Boussange
|
Victor Boussange, Bert Wuyts, Philipp Brun, Johanna T. Malle, Gabriele Midolo, Jeanne Portier, Th\'eophile Sanchez, Niklaus E. Zimmermann, Irena Axmanov\'a, Helge Bruelheide, Milan Chytr\'y, Stephan Kambach, Zde\v{n}ka Lososov\'a, Martin Ve\v{c}e\v{r}a, Idoia Biurrun, Klaus T. Ecker, Jonathan Lenoir, Jens-Christian Svenning, Dirk Nikolaus Karger
|
Multi-scale species richness estimation with deep learning
|
36 pages
| null | null | null |
q-bio.PE cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Biodiversity assessments are critically affected by the spatial scale at which species richness is measured. How species richness accumulates with sampling area depends on natural and anthropogenic processes whose effects can change depending on the spatial scale considered. These accumulation dynamics, described by the species-area relationship (SAR), are challenging to assess because most biodiversity surveys are restricted to sampling areas much smaller than the scales at which these processes operate. Here, we combine sampling theory and deep learning to predict local species richness within arbitrarily large sampling areas, enabling for the first time to estimate spatial differences in SARs. We demonstrate our approach by predicting vascular plant species richness across Europe and evaluate predictions against an independent dataset of plant community inventories. The resulting model, named deep SAR, delivers multi-scale species richness maps, improving coarse grain richness estimates by 32% compared to conventional methods, while delivering finer grain estimates. Additional to its predictive capabilities, we show how our deep SAR model can provide fundamental insights on the multi-scale effects of key biodiversity processes. The capacity of our approach to deliver comprehensive species richness estimates across the full spectrum of ecologically relevant scales is essential for robust biodiversity assessments and forecasts under global change.
|
[
{
"version": "v1",
"created": "Tue, 08 Jul 2025 19:42:33 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:43:56 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Boussange",
"Victor",
""
],
[
"Wuyts",
"Bert",
""
],
[
"Brun",
"Philipp",
""
],
[
"Malle",
"Johanna T.",
""
],
[
"Midolo",
"Gabriele",
""
],
[
"Portier",
"Jeanne",
""
],
[
"Sanchez",
"Théophile",
""
],
[
"Zimmermann",
"Niklaus E.",
""
],
[
"Axmanová",
"Irena",
""
],
[
"Bruelheide",
"Helge",
""
],
[
"Chytrý",
"Milan",
""
],
[
"Kambach",
"Stephan",
""
],
[
"Lososová",
"Zdeňka",
""
],
[
"Večeřa",
"Martin",
""
],
[
"Biurrun",
"Idoia",
""
],
[
"Ecker",
"Klaus T.",
""
],
[
"Lenoir",
"Jonathan",
""
],
[
"Svenning",
"Jens-Christian",
""
],
[
"Karger",
"Dirk Nikolaus",
""
]
] |
2507.06484
|
Fan-Yun Sun
|
Fan-Yun Sun, Shengguang Wu, Christian Jacobsen, Thomas Yim, Haoming Zou, Alex Zook, Shangru Li, Yu-Hsin Chou, Ethem Can, Xunlei Wu, Clemens Eppner, Valts Blukis, Jonathan Tremblay, Jiajun Wu, Stan Birchfield, Nick Haber
|
3D-Generalist: Self-Improving Vision-Language-Action Models for Crafting 3D Worlds
|
project website: https://ai.stanford.edu/~sunfanyun/3d-generalist/
| null | null | null |
cs.GR cs.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Despite large-scale pretraining endowing models with language and vision reasoning capabilities, improving their spatial reasoning capability remains challenging due to the lack of data grounded in the 3D world. While it is possible for humans to manually create immersive and interactive worlds through 3D graphics, as seen in applications such as VR, gaming, and robotics, this process remains highly labor-intensive. In this paper, we propose a scalable method for generating high-quality 3D environments that can serve as training data for foundation models. We recast 3D environment building as a sequential decision-making problem, employing Vision-Language-Models (VLMs) as policies that output actions to jointly craft a 3D environment's layout, materials, lighting, and assets. Our proposed framework, 3D-Generalist, trains VLMs to generate more prompt-aligned 3D environments via self-improvement fine-tuning. We demonstrate the effectiveness of 3D-Generalist and the proposed training strategy in generating simulation-ready 3D environments. Furthermore, we demonstrate its quality and scalability in synthetic data generation by pretraining a vision foundation model on the generated data. After fine-tuning the pre-trained model on downstream tasks, we show that it surpasses models pre-trained on meticulously human-crafted synthetic data and approaches results achieved with real data orders of magnitude larger.
|
[
{
"version": "v1",
"created": "Wed, 09 Jul 2025 02:00:17 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 19:36:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sun",
"Fan-Yun",
""
],
[
"Wu",
"Shengguang",
""
],
[
"Jacobsen",
"Christian",
""
],
[
"Yim",
"Thomas",
""
],
[
"Zou",
"Haoming",
""
],
[
"Zook",
"Alex",
""
],
[
"Li",
"Shangru",
""
],
[
"Chou",
"Yu-Hsin",
""
],
[
"Can",
"Ethem",
""
],
[
"Wu",
"Xunlei",
""
],
[
"Eppner",
"Clemens",
""
],
[
"Blukis",
"Valts",
""
],
[
"Tremblay",
"Jonathan",
""
],
[
"Wu",
"Jiajun",
""
],
[
"Birchfield",
"Stan",
""
],
[
"Haber",
"Nick",
""
]
] |
2507.07060
|
Shreyas Vinaya Sathyanarayana
|
Shreyas Vinaya Sathyanarayana, Sharanabasava D. Hiremath, Rahil Shah, Rishikesh Panda, Rahul Jana, Riya Singh, Rida Irfan, Ashwin Murali, Bharath Ramsundar
|
DeepRetro: Retrosynthetic Pathway Discovery using Iterative LLM Reasoning
|
64 pages,
| null | null | null |
q-bio.QM cs.AI cs.CL cs.LG q-bio.BM q-bio.MN
|
http://creativecommons.org/licenses/by/4.0/
|
The synthesis of complex natural products remains one of the grand challenges of organic chemistry. We present DeepRetro, a major advancement in computational retrosynthesis that enables the discovery of viable synthetic routes for complex molecules typically considered beyond the reach of existing retrosynthetic methods. DeepRetro is a novel, open-source framework that tightly integrates large language models (LLMs), traditional retrosynthetic engines, and expert human feedback in an iterative design loop. Prior approaches rely solely on template-based methods or unconstrained LLM outputs. In contrast, DeepRetro combines the precision of template-based methods with the generative flexibility of LLMs, controlled by rigorous chemical validity checks and enhanced by recursive refinement. This hybrid system dynamically explores and revises synthetic pathways, guided by both algorithmic checks and expert chemist feedback through an interactive user interface. While DeepRetro achieves strong performance on standard retrosynthesis benchmarks, its true strength lies in its ability to propose novel, viable pathways to highly complex natural products-targets that have historically eluded automated planning. Through detailed case studies, we illustrate how this approach enables new routes for total synthesis and facilitates human-machine collaboration in organic chemistry. Beyond retrosynthesis, DeepRetro represents a working model for how to leverage LLMs in scientific discovery. We provide a transparent account of the system's design, algorithms, and human-feedback loop, enabling broad adaptation across scientific domains. By releasing DeepRetro as an open-source tool, we aim to empower chemists to tackle increasingly ambitious synthetic targets, accelerating progress in drug discovery, materials design, and beyond.
|
[
{
"version": "v1",
"created": "Mon, 07 Jul 2025 19:41:39 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 18:39:25 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sathyanarayana",
"Shreyas Vinaya",
""
],
[
"Hiremath",
"Sharanabasava D.",
""
],
[
"Shah",
"Rahil",
""
],
[
"Panda",
"Rishikesh",
""
],
[
"Jana",
"Rahul",
""
],
[
"Singh",
"Riya",
""
],
[
"Irfan",
"Rida",
""
],
[
"Murali",
"Ashwin",
""
],
[
"Ramsundar",
"Bharath",
""
]
] |
2507.08198
|
Luke Peilen
|
David Padilla-Garza, Luke Peilen, Eric Thoma
|
Poisson Statistics for Coulomb Gases at Intermediate Temperature Regimes
|
corrected typo in abstract; minor reorganization
| null | null | null |
math.PR math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
We consider the microscopic statistics of a Coulomb gas in $\mathbb{R}^2$ at intermediate temperatures. In particular, we show that the microscopic point process associated to the Coulomb gas converges to a homogeneous Poisson point process at intermediate temperature regimes $\beta N \rightarrow \infty$ and $\beta \sqrt{N} \log N \rightarrow 0$, extending previous results. Our approach relies on a novel quantitative asymptotic description of correlation functions, which is of its own interest.
|
[
{
"version": "v1",
"created": "Thu, 10 Jul 2025 22:08:50 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 11:48:25 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Padilla-Garza",
"David",
""
],
[
"Peilen",
"Luke",
""
],
[
"Thoma",
"Eric",
""
]
] |
2507.08247
|
Peter Selinger
|
Eric Demer and Peter Selinger
|
On 3-terminal positions in Hex
|
43 pages. v2: fixed a typo in Figure 5
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper is about 3-terminal regions in Hex. A 3-terminal region is a region of the Hex board that is completely surrounded by black and white stones, in such a way that the black boundary stones form 3 connected components. We characterize Hex as the universal planar Shannon game of degree 3. This ensures that every Hex position can be decomposed into 3-terminal regions. We then investigate the combinatorial game theory of 3-terminal regions. We show that there are infinitely many distinct Hex-realizable values for such regions. We introduce an infinite family of 3-terminal positions called superswitches and investigate their properties. We also present a database of Hex-realizable 3-terminal values, and illustrate its utility as a problem-solving tool by giving various applications. The applications include the automated verification of connects-both templates and pivoting templates, a new handicap strategy for $11\times 11$ Hex, and a method for constructing witnesses for the non-inferiority of probes in many Hex templates. These methods allow us to disprove a conjecture by Henderson and Hayward.
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 01:21:07 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 04:40:39 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Demer",
"Eric",
""
],
[
"Selinger",
"Peter",
""
]
] |
2507.08377
|
Philippe Gaucher
|
Philippe Gaucher
|
Branching space of multipointed d-space
|
25 pages. companion paper of arXiv:2502.11773 ; v2: some reorganization (the numbering of the theorems is modified)
| null | null | null |
math.AT math.CT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Using the notion of short directed path, we introduce the branching space of a multipointed $d$-space. We prove that for any q-cofibrant multipointed $d$-space, it is homeomorphic to the branching space of the q-cofibrant flow obtained by applying the categorization functor. As an application, we deduce a purely topological proof of the invariance of the branching space and of the branching homology of cellular multipointed $d$-spaces up to globular subdivision. By reversing the time direction, the same results are obtained for the merging space and the merging homology.
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 07:45:29 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:44:18 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gaucher",
"Philippe",
""
]
] |
2507.08668
|
Absur Khan Siam
|
Absur Khan Siam and Se Kwon Kim
|
Chiral-split magnons in the S = 1 Shastry-Sutherland model
|
5 pages, 3 figures; fixed typos
| null | null | null |
cond-mat.mes-hall
|
http://creativecommons.org/licenses/by/4.0/
|
In ferromagnets, magnons have only one chirality; while in common antiferromagnets, bands with opposite chiralities are degenerate across the Brillouin zone. Recent studies have shown that it is possible to observe non-degenerate bands of opposite chiralities in altermagnetic materials. Here we take the S = 1 Shastry-Sutherland model, which shows the collinear N\'eel (I) phase, and investigate the magnon band structure showing alternate chirality-splitting and the resulting transport properties. In magnon bands, we find a notable feature of the chirality-split magnon bands, and the split is opposite along two different directions in the Brillouin zone. We also calculate the spin and thermal conductivities using Kubo formalism. Our calculations show robust spin Seebeck and spin Nernst effects due to the alternating chirality split across the Brillouin zone, without any external magnetic field and spin-orbit coupling.
|
[
{
"version": "v1",
"created": "Fri, 11 Jul 2025 15:08:27 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 20:37:35 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Siam",
"Absur Khan",
""
],
[
"Kim",
"Se Kwon",
""
]
] |
2507.09246
|
Bekir Can L\"utf\"uo\u{g}lu
|
Bekir Can L\"utf\"uo\u{g}lu
|
Black Holes in Proca-Gauss-Bonnet Gravity with Primary Hair: Particle Motion, Shadows, and Grey-Body Factors
|
13 pages, 8 figures, 5 tables. See https://www.sciltp.com/journals/ijgtp/articles/2507000929
|
Int. J. Grav. Theor. Phys. 1, 4 (2025)
| null | null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate classical and semiclassical signatures of black holes in a recently proposed Proca-Gauss-Bonnet gravity model that admits asymptotically flat solutions with primary hair. Two distinct classes of spherically symmetric metrics arise from different relations between the coupling constants of scalar-tensor and vector-tensor Gauss-Bonnet interactions. For each geometry, we examine the range of parameters permitting horizon formation and analyze the motion of test particles and light rays. We compute characteristic observables including the shadow radius, Lyapunov exponent, innermost stable circular orbit (ISCO) frequency, and binding energy. Additionally, we study scalar and Dirac field perturbations, derive the corresponding effective potentials, and calculate the grey-body factors (GBFs) using both the sixth-order Wentzel-Kramers-Brillouin (WKB) method and their correspondence with quasinormal modes (QNMs). Our results show that the QNM-based approximation of GBFs is accurate for sufficiently large multipole numbers and that deviations from Schwarzschild geometry become pronounced for large values of the Proca hair and Gauss-Bonnet couplings.
|
[
{
"version": "v1",
"created": "Sat, 12 Jul 2025 10:55:57 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Lütfüoğlu",
"Bekir Can",
""
]
] |
2507.09887
|
Huynh Nguyen
|
Huynh Dang Nguyen, Trong-Thang Pham, Ngan Le, Van Nguyen
|
TolerantECG: A Foundation Model for Imperfect Electrocardiogram
|
Accepted at ACM MM 2025
| null |
10.1145/3746027.3755287
| null |
cs.LG cs.AI eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
The electrocardiogram (ECG) is an essential and effective tool for diagnosing heart diseases. However, its effectiveness can be compromised by noise or unavailability of one or more leads of the standard 12-lead recordings, resulting in diagnostic errors or uncertainty. To address these challenges, we propose TolerantECG, a foundation model for ECG signals that is robust to noise and capable of functioning with arbitrary subsets of the standard 12-lead ECG. TolerantECG training combines contrastive and self-supervised learning frameworks to jointly learn ECG signal representations alongside their corresponding knowledge-retrieval-based text report descriptions and corrupted or lead-missing signals. Comprehensive benchmarking results demonstrate that TolerantECG consistently ranks as the best or second-best performer across various ECG signal conditions and class levels in the PTB-XL dataset, and achieves the highest performance on the MIT-BIH Arrhythmia Database.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 03:48:35 GMT"
},
{
"version": "v2",
"created": "Tue, 29 Jul 2025 06:12:37 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 08:07:02 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Nguyen",
"Huynh Dang",
""
],
[
"Pham",
"Trong-Thang",
""
],
[
"Le",
"Ngan",
""
],
[
"Nguyen",
"Van",
""
]
] |
2507.10016
|
Lixu Wang
|
Lixu Wang, Kaixiang Yao, Xinfeng Li, Dong Yang, Haoyang Li, Xiaofeng Wang, Wei Dong
|
The Man Behind the Sound: Demystifying Audio Private Attribute Profiling via Multimodal Large Language Model Agents
|
22 pages, 4 figures
| null | null | null |
cs.CR cs.SD eess.AS
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Our research uncovers a novel privacy risk associated with multimodal large language models (MLLMs): the ability to infer sensitive personal attributes from audio data -- a technique we term audio private attribute profiling. This capability poses a significant threat, as audio can be covertly captured without direct interaction or visibility. Moreover, compared to images and text, audio carries unique characteristics, such as tone and pitch, which can be exploited for more detailed profiling. However, two key challenges exist in understanding MLLM-employed private attribute profiling from audio: (1) the lack of audio benchmark datasets with sensitive attribute annotations and (2) the limited ability of current MLLMs to infer such attributes directly from audio. To address these challenges, we introduce AP^2, an audio benchmark dataset that consists of two subsets collected and composed from real-world data, and both are annotated with sensitive attribute labels. Additionally, we propose Gifts, a hybrid multi-agent framework that leverages the complementary strengths of audio-language models (ALMs) and large language models (LLMs) to enhance inference capabilities. Gifts employs an LLM to guide the ALM in inferring sensitive attributes, then forensically analyzes and consolidates the ALM's inferences, overcoming severe hallucinations of existing ALMs in generating long-context responses. Our evaluations demonstrate that Gifts significantly outperforms baseline approaches in inferring sensitive attributes. Finally, we investigate model-level and data-level defense strategies to mitigate the risks of audio private attribute profiling. Our work validates the feasibility of audio-based privacy attacks using MLLMs, highlighting the need for robust defenses, and provides a dataset and framework to facilitate future research.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 07:51:56 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:04:41 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wang",
"Lixu",
""
],
[
"Yao",
"Kaixiang",
""
],
[
"Li",
"Xinfeng",
""
],
[
"Yang",
"Dong",
""
],
[
"Li",
"Haoyang",
""
],
[
"Wang",
"Xiaofeng",
""
],
[
"Dong",
"Wei",
""
]
] |
2507.10348
|
Yichen Li
|
Yichen Li, Xiuying Wang, Wenchao Xu, Haozhao Wang, Yining Qi, Jiahua Dong, Ruixuan Li
|
Feature Distillation is the Better Choice for Model-Heterogeneous Federated Learning
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Model-Heterogeneous Federated Learning (Hetero-FL) has attracted growing attention for its ability to aggregate knowledge from heterogeneous models while keeping private data locally. To better aggregate knowledge from clients, ensemble distillation, as a widely used and effective technique, is often employed after global aggregation to enhance the performance of the global model. However, simply combining Hetero-FL and ensemble distillation does not always yield promising results and can make the training process unstable. The reason is that existing methods primarily focus on logit distillation, which, while being model-agnostic with softmax predictions, fails to compensate for the knowledge bias arising from heterogeneous models. To tackle this challenge, we propose a stable and efficient Feature Distillation for model-heterogeneous Federated learning, dubbed FedFD, that can incorporate aligned feature information via orthogonal projection to integrate knowledge from heterogeneous models better. Specifically, a new feature-based ensemble federated knowledge distillation paradigm is proposed. The global model on the server needs to maintain a projection layer for each client-side model architecture to align the features separately. Orthogonal techniques are employed to re-parameterize the projection layer to mitigate knowledge bias from heterogeneous models and thus maximize the distilled knowledge. Extensive experiments show that FedFD achieves superior performance compared to state-of-the-art methods.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 14:51:18 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 06:59:30 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Li",
"Yichen",
""
],
[
"Wang",
"Xiuying",
""
],
[
"Xu",
"Wenchao",
""
],
[
"Wang",
"Haozhao",
""
],
[
"Qi",
"Yining",
""
],
[
"Dong",
"Jiahua",
""
],
[
"Li",
"Ruixuan",
""
]
] |
2507.10489
|
Eduardo Ribas Brito
|
Eduardo Brito, Mahmoud Shoush, Kristian Tamm, Paula Etti and Liina Kamm
|
SynthGuard: Redefining Synthetic Data Generation with a Scalable and Privacy-Preserving Workflow Framework
|
This is the extended version of the paper to appear in the Proceedings of the 1st International Workshop on Responsible Data Governance, Privacy, and Digital Transformation (RDGPT 2025), held in conjunction with the 20th International Conference on Availability, Reliability and Security (ARES 2025)
| null |
10.1007/978-3-032-00633-2_12
| null |
cs.CR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The growing reliance on data-driven applications in sectors such as healthcare, finance, and law enforcement underscores the need for secure, privacy-preserving, and scalable mechanisms for data generation and sharing. Synthetic data generation (SDG) has emerged as a promising approach but often relies on centralized or external processing, raising concerns about data sovereignty, domain ownership, and compliance with evolving regulatory standards. To overcome these issues, we introduce SynthGuard, a framework designed to ensure computational governance by enabling data owners to maintain control over SDG workflows. SynthGuard supports modular and privacy-preserving workflows, ensuring secure, auditable, and reproducible execution across diverse environments. In this paper, we demonstrate how SynthGuard addresses the complexities at the intersection of domain-specific needs and scalable SDG by aligning with requirements for data sovereignty and regulatory compliance. Developed iteratively with domain expert input, SynthGuard has been validated through real-world use cases, demonstrating its ability to balance security, privacy, and scalability while ensuring compliance. The evaluation confirms its effectiveness in implementing and executing SDG workflows and integrating privacy and utility assessments across various computational environments.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 17:11:20 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Brito",
"Eduardo",
""
],
[
"Shoush",
"Mahmoud",
""
],
[
"Tamm",
"Kristian",
""
],
[
"Etti",
"Paula",
""
],
[
"Kamm",
"Liina",
""
]
] |
2507.10772
|
Michal Podstawski
|
Michal Podstawski
|
Applying Text Embedding Models for Efficient Analysis in Labeled Property Graphs
| null | null | null | null |
cs.CL cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
Labeled property graphs often contain rich textual attributes that can enhance analytical tasks when properly leveraged. This work explores the use of pretrained text embedding models to enable efficient semantic analysis in such graphs. By embedding textual node and edge properties, we support downstream tasks including node classification and relation prediction with improved contextual understanding. Our approach integrates language model embeddings into the graph pipeline without altering its structure, demonstrating that textual semantics can significantly enhance the accuracy and interpretability of property graph analysis.
|
[
{
"version": "v1",
"created": "Mon, 14 Jul 2025 19:53:56 GMT"
},
{
"version": "v2",
"created": "Wed, 13 Aug 2025 10:27:38 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 13:04:34 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Podstawski",
"Michal",
""
]
] |
2507.11231
|
Aditya Sow Mondal
|
Aditya S. Mondal, Mayukh Pahari, Gulab C. Dewangan
|
NuSTAR discovers a long type-I X-ray burst from the clocked burster GS 1826-24
|
19 pages, 13 figures, 2 tables. Accepted for publication in the Journal of High Energy Astrophysics (JHEAP)
| null | null | null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The source GS~1826-24 is a neutron star low mass X-ray binary known as the 'clocked burster' because of its extremely regular bursting behavior. We report on the detection of a long type-I X-ray burst from this source. We perform a detailed spectroscopic analysis of the long X-ray burst, lasting for $\sim 600$ s, seen in the NuSTAR observation carried out on 2022 September. The persistent emission is well described by an absorbed thermal Comptonization model (nthcomp), and the source exhibits a soft spectral state during this observation. The observed burst exhibits a rise time of $\sim 25$ s and a decay time of $\sim 282$ s. The time-resolved spectroscopy of the burst shows a significant departure from a pure thermal spectrum and is described with a model consisting of a varying-temperature blackbody plus an evolving persistent emission component. We observe a significant enhancement in the persistent emission during the burst. The enhancement of the pre-burst persistent flux is possibly due to Poynting-Robertson drag or coronal reprocessing. At the peak of the burst, the blackbody temperature and the blackbody emitting radius reached a maximum of $2.10\pm 0.07$ keV and $5.5\pm 2.1$ km, respectively. The peak flux ($F_{peak}$) during the burst is $\approx 2.4\times 10^{-8}$ ergs cm$^{-2}$ s$^{-1}$, which corresponds to a luminosity of $\approx 9.7\times 10^{37}$ ergs s$^{-1}$.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 12:02:12 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:10:10 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Mondal",
"Aditya S.",
""
],
[
"Pahari",
"Mayukh",
""
],
[
"Dewangan",
"Gulab C.",
""
]
] |
2507.11237
|
Kathleen Sammut
|
Kathleen Sammut
|
Testing Planck 2020 and DESI data on wCDM Models
|
16 figures, 36 pages, 19 tables
| null | null | null |
astro-ph.CO gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
This work delves into the dynamical dark energy models of the wCDM parameterisation that are defined by their equation of state by comparing different, well-known parameterisation models, in an attempt to lessen the tensions of {H_0} and {\sigma_8} by using the latest observational data. This research also tested the newer Planck likelihood and compared it to the previously released dataset of Planck 2018 and the newer BAO data of DESI. The data that was used were: the Cosmic Microwave Background (CMB) data of Planck 2018 and Planck 2020 data; Cosmic Chronometers (CC), a sample of Supernovae Type Ia; and Baryonic Acoustic Oscillations (BAO). A Bayesian analysis was performed to produce the results needed for the analysis. From the analyses, the best-fit values of the parameters show that almost all models are in favour of a phantom Universe when early-time data was used and favours a quintessence Universe when combinations of early-time and late-time data was considered. This study also showed that the new tested data constrained the models better than the previous ones, while also showing that the observation data supports the {\Lambda}CDM model over the dynamical dark energy models
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 12:11:56 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 15:39:39 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 10:51:55 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sammut",
"Kathleen",
""
]
] |
2507.11317
|
Darukesha Baraduru Hirematada
|
Darukesha B H M
|
Species-Dependent Electron Emission from Nanoparticles under Gamma Irradiation
|
Version 2 (Uncertainty and Error Analysis are brought together in one place). Comments are welcome
| null | null | null |
physics.med-ph physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this study, various nanoparticle species-including Au and Gd$_2$O$_3$-were irradiated with low-energy gamma rays, such as 59.5 keV photons from $^{241}$Am. Pulse-height spectra were recorded using a liquid-scintillation counting system before and after dispersing the nanoparticles into the scintillator, and the differences between them were analyzed to infer the interaction outcomes. Gd$_2$O$_3$ nanoparticles emitted numerous electrons; however, under identical experimental conditions, no detectable electron emission was observed from Au nanoparticles (AuNPs). Here, "detectable electron emission" refers to electrons with energies high enough to be registered by the liquid-scintillation detector used (approx 100 eV and, more typically, >= 1-2 keV); however, it excludes electrons that may be emitted at lower energies. Thus, a species-dependent radiation-nanoparticle interaction was observed. Rigorous controls and falsification tests excluded various artefacts-including detector insensitivity, surface contamination, aggregation, quenching, and self-absorption-as causes for the absence of detectable electron emission from AuNPs. This observation potentially prompts a re-evaluation of the common assumption that nanoparticles behave like their bulk counterparts and emit electrons upon gamma-irradiation. Instead, our results suggest that the distinct internal environment of nanoparticles influences their interaction with radiation. These findings offer significant insights for practical applications, including a better mechanistic understanding of nanoparticle radiosensitization in cancer therapy, enhanced gamma-detection efficiency of organic scintillators, and the development of lightweight radiation shields.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 13:49:03 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:32:11 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"M",
"Darukesha B H",
""
]
] |
2507.11579
|
Sathvik Chereddy
|
Sathvik Chereddy and John Femiani
|
SketchDNN: Joint Continuous-Discrete Diffusion for CAD Sketch Generation
|
17 pages, 63 figures, Proceedings of the 42nd International Conference on Machine Learning (ICML2025)
| null | null | null |
cs.CV cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We present SketchDNN, a generative model for synthesizing CAD sketches that jointly models both continuous parameters and discrete class labels through a unified continuous-discrete diffusion process. Our core innovation is Gaussian-Softmax diffusion, where logits perturbed with Gaussian noise are projected onto the probability simplex via a softmax transformation, facilitating blended class labels for discrete variables. This formulation addresses 2 key challenges, namely, the heterogeneity of primitive parameterizations and the permutation invariance of primitives in CAD sketches. Our approach significantly improves generation quality, reducing Fr\'echet Inception Distance (FID) from 16.04 to 7.80 and negative log-likelihood (NLL) from 84.8 to 81.33, establishing a new state-of-the-art in CAD sketch generation on the SketchGraphs dataset.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 08:27:31 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 06:40:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chereddy",
"Sathvik",
""
],
[
"Femiani",
"John",
""
]
] |
2507.11643
|
Vladimir Kanovei
|
Vladimir Kanovei and Vassily Lyubetsky
|
Notes on the equiconsistency of ZFC without the Power Set axiom and 2nd order PA
|
Version 2 revised and updated
| null | null | null |
math.LO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We demonstrate that theories $\text{Z}^-$, $\text{ZF}^-$, $\text{ZFC}^-$ (minus means the absence of the Power Set axiom) and $\text{PA}_2$, $\text{PA}_2^-$ (minus means the absence of the Countable Choice schema) are equiconsistent to each other. The methods used include the interpretation of a power-less set theory in $\text{PA}_2^-$ via well-founded extensional digraphs, as well as the G\"odel constructibility in the said power-less set theory.
|
[
{
"version": "v1",
"created": "Tue, 15 Jul 2025 18:28:26 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 01:57:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kanovei",
"Vladimir",
""
],
[
"Lyubetsky",
"Vassily",
""
]
] |
2507.12424
|
Michael Potter
|
Michael Potter and Michael Everett and Deniz Erdogmus and Yuna Watanabe and Tales Imbiriba and Matthew S. Goodwin
|
Hierarchical Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youth with Autism for Branching Factor Estimation
|
Submitted to BMC Medical Research Methodology
| null | null | null |
stat.AP math.PR stat.OT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Aggressive behavior in autistic inpatient youth often arises in temporally clustered bursts complicating efforts to distinguish external triggers from internal escalation. The sample population branching factor-the expected number of new onsets triggered by a given event-is a key summary of self-excitation in behavior dynamics. Prior pooled models overestimate this quantity by ignoring patient-specific variability. We addressed this using a hierarchical Hawkes process with an exponential kernel and edge-effect correction allowing partial pooling across patients. This approach reduces bias from high-frequency individuals and stabilizes estimates for those with sparse data. Bayesian inference was performed using the No U-Turn Sampler with model evaluation via convergence diagnostics, power-scaling sensitivity analysis, and multiple Goodness-of-Fit (GOF) metrics: PSIS-LOO the Lewis test with Durbin's modification and residual analysis based on the Random Time Change Theorem (RTCT). The hierarchical model yielded a significantly lower and more precise branching factor estimate mean (0.742 +- 0.026) than the pooled model (0.899 +- 0.015) and narrower intervals than the unpooled model (0.717 +- 0.139). This led to a threefold smaller cascade of events per onset under the hierarchical model. Sensitivity analyses confirmed robustness to prior and likelihood perturbations while the unpooled model showed instability for sparse individuals. GOF measures consistently favored or on par to the hierarchical model. Hierarchical Hawkes modeling with edge-effect correction provides robust estimation of branching dynamics by capturing both within- and between-patient variability. This enables clearer separation of endogenous from exogenous events supports linkage to physiological signals and enhances early warning systems individualized treatment and resource allocation in inpatient care.
|
[
{
"version": "v1",
"created": "Wed, 16 Jul 2025 17:11:48 GMT"
},
{
"version": "v2",
"created": "Sun, 20 Jul 2025 23:14:00 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 21:19:02 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Potter",
"Michael",
""
],
[
"Everett",
"Michael",
""
],
[
"Erdogmus",
"Deniz",
""
],
[
"Watanabe",
"Yuna",
""
],
[
"Imbiriba",
"Tales",
""
],
[
"Goodwin",
"Matthew S.",
""
]
] |
2507.12911
|
Hayeon Oh
|
Hayeon Oh
|
LaViPlan : Language-Guided Visual Path Planning with RLVR
|
Accepted to the 2nd ICCV 2025 Workshop on the Challenge of Out-of-Label Hazards in Autonomous Driving (13 pages, 6 figures)
| null | null | null |
cs.RO cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Out-of-distribution (OOD) scenarios in autonomous driving pose critical challenges, as planners often fail to generalize beyond their training experience, leading to unsafe or unexpected behavior. Vision-Language Models (VLMs) have shown promise in handling such scenarios by providing high-level scene understanding and user-aligned decisions. However, existing VLMs often exhibit a misalignment between their language-based reasoning and the low-level trajectories required for action-level planning. In this paper, we propose LaViPlan, a framework that leverages Reinforcement Learning with Verifiable Rewards (RLVR) to fine-tune VLMs using planning-oriented metrics. Experimental results show that LaViPlan improves planning performance across both in-domain and out-of-domain datasets. While linguistic fidelity slightly decreases after RLVR-based fine-tuning, qualitative evaluation indicates that the outputs remain coherent. We also conduct ablation studies to analyze the effects of sampling ratio and reasoning guidance, highlighting how these design choices influence performance. These findings demonstrate the potential of RLVR as a post-training paradigm for aligning language-guided reasoning with action-level planning in autonomous driving.
|
[
{
"version": "v1",
"created": "Thu, 17 Jul 2025 08:58:24 GMT"
},
{
"version": "v2",
"created": "Mon, 21 Jul 2025 01:01:29 GMT"
},
{
"version": "v3",
"created": "Wed, 23 Jul 2025 01:28:44 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 06:32:37 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Oh",
"Hayeon",
""
]
] |
2507.13306
|
Zhenkang Lu
|
Zhenkang Lu, Cheng Ran, Shao-feng Wu
|
The Algebraic Structure Underlying Pole-Skipping Points
|
49 pages, 4 figures, v2: minor corrections
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The holographic Green's function becomes ambiguous, taking the indeterminate form `$0/0$', at an infinite set of special frequencies and momenta known as ``pole-skipping points''. In this work, we propose that these pole-skipping points can be used to reconstruct both the interior and exterior geometry of a static, planar-symmetric black hole in the bulk. The entire reconstruction procedure is fully analytical and only involves solving a system of linear equations. We demonstrate its effectiveness across various backgrounds, including the BTZ black hole, its $T\bar{T}$-deformed counterparts, as well as geometries with Lifshitz scaling and hyperscaling-violation. Within this framework, other geometric quantities, such as the vacuum Einstein equations, can also be reinterpreted directly in terms of pole-skipping data. Moreover, our approach reveals a hidden algebraic structure governing the pole-skipping points of Klein-Gordon equations of the form $(\nabla^{2} + V(r))\phi(r) = 0$: only a subset of these points is independent, while the remainder is constrained by an equal number of homogeneous polynomial identities in the pole-skipping momenta. These identities are universal, as confirmed by their validity across a broad class of bulk geometries with varying dimensionality, boundary asymptotics, and perturbation modes.
|
[
{
"version": "v1",
"created": "Thu, 17 Jul 2025 17:22:41 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 17:47:56 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Lu",
"Zhenkang",
""
],
[
"Ran",
"Cheng",
""
],
[
"Wu",
"Shao-feng",
""
]
] |
2507.14111
|
Jiwei Li
|
Xiaoya Li, Xiaofei Sun, Albert Wang, Jiwei Li and Chris Shum
|
CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning
|
Project Page: https://deepreinforce-ai.github.io/cudal1_blog/
| null | null | null |
cs.AI cs.DC cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
The exponential growth in demand for GPU computing resources has created an urgent need for automated CUDA optimization strategies. While recent advances in LLMs show promise for code generation, current SOTA models achieve low success rates in improving CUDA speed. In this paper, we introduce CUDA-L1, an automated reinforcement learning framework for CUDA optimization that employs a novel contrastive RL algorithm.
CUDA-L1 achieves significant performance improvements on the CUDA optimization task: trained on A100, it delivers an average speedup of x3.12 with a median speedup of x1.42 against default baselines over across all 250 CUDA kernels of KernelBench, with peak speedups reaching x120. In addition to the default baseline provided by KernelBench, CUDA-L1 demonstrates x2.77 over Torch Compile, x2.88 over Torch Compile with reduce overhead, x2.81 over CUDA Graph implementations, and remarkably x7.72 over cuDNN libraries. Furthermore, the model also demonstrates portability across different GPU architectures.
Beyond these benchmark results, CUDA-L1 demonstrates several properties: it 1) discovers a variety of CUDA optimization techniques and learns to combine them strategically to achieve optimal performance; 2) uncovers fundamental principles of CUDA optimization, such as the multiplicative nature of optimizations; 3) identifies non-obvious performance bottlenecks and rejects seemingly beneficial optimizations that actually harm performance. The capabilities demonstrate that, RL can transform an initially poor-performing LLM into an effective CUDA optimizer through speedup-based reward signals alone, without human expertise or domain knowledge. This paradigm opens possibilities for automated optimization of CUDA operations, and holds promise to substantially promote GPU efficiency and alleviate the rising pressure on GPU computing resources.
|
[
{
"version": "v1",
"created": "Fri, 18 Jul 2025 17:43:56 GMT"
},
{
"version": "v2",
"created": "Mon, 21 Jul 2025 07:28:22 GMT"
},
{
"version": "v3",
"created": "Tue, 22 Jul 2025 07:42:18 GMT"
},
{
"version": "v4",
"created": "Mon, 28 Jul 2025 07:04:11 GMT"
},
{
"version": "v5",
"created": "Sat, 02 Aug 2025 01:46:46 GMT"
},
{
"version": "v6",
"created": "Tue, 12 Aug 2025 00:12:38 GMT"
},
{
"version": "v7",
"created": "Wed, 20 Aug 2025 01:41:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Li",
"Xiaoya",
""
],
[
"Sun",
"Xiaofei",
""
],
[
"Wang",
"Albert",
""
],
[
"Li",
"Jiwei",
""
],
[
"Shum",
"Chris",
""
]
] |
2507.14262
|
Arpit Singh
|
Deependra Sharma, Arpit Singh, Md. Samsul Islam, Sadhana Dash
|
Deciphering the dynamics of nuclear collisions with elongated structure of $^{20}$Ne
|
6 pages, 3 figures
| null | null | null |
nucl-th hep-ph hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We present the results obtained from the study of $^{20}$Ne$-$$^{20}$Ne collisions at $\sqrt{s_{NN}}$= 6.8 TeV using the Monte Carlo Pythia8/Angantyr model. Angantyr models heavy-ion collisions as a superposition of independent nucleon-nucleon (NN) collisions, without incorporating collective effects. We construct a bi-pyramidal structure of $^{20}$Ne nucleus composed of five $\alpha$-clusters and compare its collision dynamics against those generated using Woods-Saxon nuclear density distribution. The results are further compared for different orientations of the $^{20}$Ne nuclei, including tip-tip, body-body, body-tip, and random orientations of the bi-pyramidal structure. The results show that the geometric arrangement of $\alpha$-clusters in the $^{20}$Ne nucleus significantly influences the particle production patterns in final state particle multiplicity and mean transverse momentum ($\langle p_T \rangle$). These results highlight the sensitivity of final state observables to the nuclear structure and orientation of colliding nuclei, providing insights into the dynamics of small collision systems, even in non-hydrodynamic framework.
|
[
{
"version": "v1",
"created": "Fri, 18 Jul 2025 13:25:41 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:09:05 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sharma",
"Deependra",
""
],
[
"Singh",
"Arpit",
""
],
[
"Islam",
"Md. Samsul",
""
],
[
"Dash",
"Sadhana",
""
]
] |
2507.14289
|
Kaloyan Penev
|
Kaloyan Penev, Angel Romero, S. Javad Jafarzadeh, Olivier Guyon, Wilfred Gee, Preethi Krishnamoorthy
|
Tools for High Precision Photometry from Wide-Field Color Images
|
Accepted for publication in AJ, 9 pages, 5 figures
| null |
10.3847/1538-3881/adf640
| null |
astro-ph.IM astro-ph.SR
|
http://creativecommons.org/licenses/by/4.0/
|
We present AstroWISP: a collection of image processing tools for source extraction, background determination, point spread function/pixel response function fitting, and aperture photometry. AstroWISP is particularly well-suited for working with detectors featuring a Bayer mask (an array of microfilters applied to each detector pixel to allow color photography), such as consumer DSLR cameras. Such detectors pose significant challenges for existing tools while offering a much cheaper alternative to specialized devices. As a result, consumer DSLR cameras with Bayer masks are often underutilized for precision photometry. \astrowisp{} addresses this limitation in an effort to democratize precision photometry and support broader community participation in research. We demonstrate that our tools produce high-precision photometry from such images, enabling the use of such devices for detecting exoplanet transits. We package our tools for all major operating systems to ensure accessibility for amateur astronomers.
|
[
{
"version": "v1",
"created": "Fri, 18 Jul 2025 18:00:49 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:25:34 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Penev",
"Kaloyan",
""
],
[
"Romero",
"Angel",
""
],
[
"Jafarzadeh",
"S. Javad",
""
],
[
"Guyon",
"Olivier",
""
],
[
"Gee",
"Wilfred",
""
],
[
"Krishnamoorthy",
"Preethi",
""
]
] |
2507.15431
|
Andrew Gracyk
|
Andrew Gracyk
|
The calculus of variations of the Transformer on the hyperspherical tangent bundle
|
Small corrections and improvements, primarily to Theorem 4
| null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We offer a theoretical mathematical background to Transformers through Lagrangian optimization across the token space. The Transformer, as a flow map, exists in the tangent fiber for each token along the high-dimensional unit sphere. The circumstance of the hypersphere across the latent data is reasonable due to the trained diagonal matrix equal to the identity, which has various empirical justifications. Thus, under the continuum limit of the dynamics, the latent vectors flow among the tangent bundle. Using these facts, we devise a mathematical framework for the Transformer through calculus of variations. We develop a functional and show that the continuous flow map induced by the Transformer satisfies this functional, therefore the Transformer can be viewed as a natural solver of a calculus of variations problem. We invent new scenarios of when our methods are applicable based on loss optimization with respect to path optimality. We derive the Euler-Lagrange equation for the Transformer. The variant of the Euler-Lagrange equation we present has various appearances in literature, but, to our understanding, oftentimes not foundationally proven or under other specialized cases. Our overarching proof is new: our techniques are classical and the use of the flow map object is original. We provide several other relevant results, primarily ones specific to neural scenarios. In particular, much of our analysis will be attempting to quantify Transformer data in variational contexts under neural approximations. Calculus of variations on manifolds is a well-nourished research area, but for the Transformer specifically, it is uncharted: we lay the foundation for this area through an introduction to the Lagrangian for the Transformer.
|
[
{
"version": "v1",
"created": "Mon, 21 Jul 2025 09:43:33 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 20:14:43 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gracyk",
"Andrew",
""
]
] |
2507.15453
|
Yong-Fan Chen
|
Po-Han Tseng and Yong-Fan Chen
|
Entanglement Preservation and Clauser-Horne Nonlocality in Electromagnetically Induced Transparency Quantum Memories
|
17 pages, 4 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Quantum memories are indispensable for quantum repeater networks, deterministic generation of single- or multi-photon states, and linear optical quantum computing. Although experiments have demonstrated that electromagnetically induced transparency (EIT) quantum memories can store entangled photons, a definitive theoretical framework verifying the preservation of entanglement and nonlocality has not been rigorously established. Here, we develop a comprehensive model by integrating the dark-state polariton framework with reduced-density-operator theory to derive the retrieved density operator under realistic ground-state decoherence conditions. Our analysis reveals that decoherence inevitably causes probe photon loss, converting a Bell state into a mixed state. Crucially, we predict a critical storage efficiency threshold of 89.7%. Only when this threshold is exceeded does the retrieved probe photon state violate the Clauser-Horne inequality, thereby preserving nonlocality; below this point, nonlocal correlations vanish. Moreover, our theory shows that multiple spatially separated EIT memories can cooperatively store and retrieve N-qubit entangled states encoded in photon number, path, and polarization with near-unity fidelity in the ideal limit. This work bridges a long-standing theoretical gap and lays a solid foundation for the application of EIT memories in scalable quantum networks and quantum information processing.
|
[
{
"version": "v1",
"created": "Mon, 21 Jul 2025 10:05:11 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:40:35 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tseng",
"Po-Han",
""
],
[
"Chen",
"Yong-Fan",
""
]
] |
2507.16191
|
Xiantao Hu
|
Fansheng Zeng and Bineng Zhong and Haiying Xia and Yufei Tan and Xiantao Hu and Liangtao Shi and Shuxiang Song
|
Explicit Context Reasoning with Supervision for Visual Tracking
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Contextual reasoning with constraints is crucial for enhancing temporal consistency in cross-frame modeling for visual tracking. However, mainstream tracking algorithms typically associate context by merely stacking historical information without explicitly supervising the association process, making it difficult to effectively model the target's evolving dynamics. To alleviate this problem, we propose RSTrack, which explicitly models and supervises context reasoning via three core mechanisms. \textit{1) Context Reasoning Mechanism}: Constructs a target state reasoning pipeline, converting unconstrained contextual associations into a temporal reasoning process that predicts the current representation based on historical target states, thereby enhancing temporal consistency. \textit{2) Forward Supervision Strategy}: Utilizes true target features as anchors to constrain the reasoning pipeline, guiding the predicted output toward the true target distribution and suppressing drift in the context reasoning process. \textit{3) Efficient State Modeling}: Employs a compression-reconstruction mechanism to extract the core features of the target, removing redundant information across frames and preventing ineffective contextual associations. These three mechanisms collaborate to effectively alleviate the issue of contextual association divergence in traditional temporal modeling. Experimental results show that RSTrack achieves state-of-the-art performance on multiple benchmark datasets while maintaining real-time running speeds. Our code is available at https://github.com/GXNU-ZhongLab/RSTrack.
|
[
{
"version": "v1",
"created": "Tue, 22 Jul 2025 03:07:50 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 02:43:14 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zeng",
"Fansheng",
""
],
[
"Zhong",
"Bineng",
""
],
[
"Xia",
"Haiying",
""
],
[
"Tan",
"Yufei",
""
],
[
"Hu",
"Xiantao",
""
],
[
"Shi",
"Liangtao",
""
],
[
"Song",
"Shuxiang",
""
]
] |
2507.17005
|
Roudy Hanna
|
Roudy Hanna, S\"oren Ihssen, Simon Geisert, Umut Kocak, Matteo Arfini, Albert Hertel, Thomas J. Smart, Michael Schleenvoigt, Tobias Schmitt, Joscha Domnick, Kaycee Underwood, Abdur Rehman Jalil, Jin Hee Bae, Benjamin Bennemann, Mathieu F\'echant, Mitchell Field, Martin Spiecker, Nicolas Zapata, Christian Dickel, Erwin Berenschot, Niels Tas, Gary A. Steele, Detlev Gr\"utzmacher, Ioan M. Pop, Peter Sch\"uffelgen
|
On-chip stencil lithography for superconducting qubits
|
v1: 14 pages, 13 figures, v2: typos corrected
| null | null | null |
quant-ph
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Improvements in circuit design and more recently in materials and surface cleaning have contributed to a rapid development of coherent superconducting qubits. However, organic resists commonly used for shadow evaporation of Josephson junctions (JJs) pose limitations due to residual contamination, poor thermal stability and compatibility under typical surface-cleaning conditions. To provide an alternative, we developed an inorganic SiO$_2$/Si$_3$N$_4$ on-chip stencil lithography mask for JJ fabrication. The stencil mask is resilient to aggressive cleaning agents and it withstands high temperatures up to 1200{\deg}C, thereby opening new avenues for JJ material exploration and interface optimization. To validate the concept, we performed shadow evaporation of Al-based transmon qubits followed by stencil mask lift-off using vapor hydrofluoric acid, which selectively etches SiO$_2$. We demonstrate average $T_1 \approx 75 \pm 11 \mu$s over a 200 MHz frequency range in multiple cool-downs for one device, and $T_1 \approx 44\pm 8 \mu$s for a second device. These results confirm the compatibility of stencil lithography with state-of-the-art superconducting quantum devices and motivate further investigations into materials engineering, film deposition and surface cleaning techniques.
|
[
{
"version": "v1",
"created": "Tue, 22 Jul 2025 20:31:21 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:03:12 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hanna",
"Roudy",
""
],
[
"Ihssen",
"Sören",
""
],
[
"Geisert",
"Simon",
""
],
[
"Kocak",
"Umut",
""
],
[
"Arfini",
"Matteo",
""
],
[
"Hertel",
"Albert",
""
],
[
"Smart",
"Thomas J.",
""
],
[
"Schleenvoigt",
"Michael",
""
],
[
"Schmitt",
"Tobias",
""
],
[
"Domnick",
"Joscha",
""
],
[
"Underwood",
"Kaycee",
""
],
[
"Jalil",
"Abdur Rehman",
""
],
[
"Bae",
"Jin Hee",
""
],
[
"Bennemann",
"Benjamin",
""
],
[
"Féchant",
"Mathieu",
""
],
[
"Field",
"Mitchell",
""
],
[
"Spiecker",
"Martin",
""
],
[
"Zapata",
"Nicolas",
""
],
[
"Dickel",
"Christian",
""
],
[
"Berenschot",
"Erwin",
""
],
[
"Tas",
"Niels",
""
],
[
"Steele",
"Gary A.",
""
],
[
"Grützmacher",
"Detlev",
""
],
[
"Pop",
"Ioan M.",
""
],
[
"Schüffelgen",
"Peter",
""
]
] |
2507.17008
|
Gaston Gustavo Rios
|
Gaston Gustavo Rios, Pedro Dal Bianco, Franco Ronchetti, Facundo Quiroga, Oscar Stanchi, Santiago Ponte Ah\'on, Waldo Hasperu\'e
|
Bringing Balance to Hand Shape Classification: Mitigating Data Imbalance Through Generative Models
|
23 pages, 8 figures, to be published in Applied Soft Computing
|
Applied Soft Computing (2025)
|
10.1016/j.asoc.2025.113586
| null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Most sign language handshape datasets are severely limited and unbalanced, posing significant challenges to effective model training. In this paper, we explore the effectiveness of augmenting the training data of a handshape classifier by generating synthetic data. We use an EfficientNet classifier trained on the RWTH German sign language handshape dataset, which is small and heavily unbalanced, applying different strategies to combine generated and real images. We compare two Generative Adversarial Networks (GAN) architectures for data generation: ReACGAN, which uses label information to condition the data generation process through an auxiliary classifier, and SPADE, which utilizes spatially-adaptive normalization to condition the generation on pose information. ReACGAN allows for the generation of realistic images that align with specific handshape labels, while SPADE focuses on generating images with accurate spatial handshape configurations. Our proposed techniques improve the current state-of-the-art accuracy on the RWTH dataset by 5%, addressing the limitations of small and unbalanced datasets. Additionally, our method demonstrates the capability to generalize across different sign language datasets by leveraging pose-based generation trained on the extensive HaGRID dataset. We achieve comparable performance to single-source trained classifiers without the need for retraining the generator.
|
[
{
"version": "v1",
"created": "Tue, 22 Jul 2025 20:41:29 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Rios",
"Gaston Gustavo",
""
],
[
"Bianco",
"Pedro Dal",
""
],
[
"Ronchetti",
"Franco",
""
],
[
"Quiroga",
"Facundo",
""
],
[
"Stanchi",
"Oscar",
""
],
[
"Ahón",
"Santiago Ponte",
""
],
[
"Hasperué",
"Waldo",
""
]
] |
2507.17238
|
Kartik Panda
|
Kartik Panda, Itzik Kapon, Dumitru Dumcenco, Dirk van der Marel, Alexey Kuzmenko, and Nimrod Bachar
|
Coupling of magnetic and lattice collective excitations in the 2D van der Waals antiferromagnet FePS$_{3}$
|
17 pages, 15 figures
| null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
We combine polarized infrared magneto-transmission and Faraday angle rotation measurements to map the collective excitations of the van der Waals antiferromagnet FePS$_3$. Below the N\'{e}el temperature ($T_\mathrm{N} \approx 118$ K), the phonon spectrum becomes strongly anisotropic, while a prominent excitation at 122 cm$^{-1}$ (15 meV) hardens on cooling and splits linearly with magnetic field,identifying its magnetic origin. From absolute transmission and Faraday angle measurements, we reconstruct the circular optical conductivities, revealing a pronounced dichroism of the field-split excitations. The upper branch near 129 cm$^{-1}$ shows an anomalously reduced dichroism, consistent with spin-phonon hybridization with a nearby infrared phonon. Several phonons exhibit sizable Faraday rotation, and Raman-active modes appear in the infrared spectra. Together with a broad mid-infrared band near 900 cm$^{-1}$ emerging only below $T_\mathrm{N}$, these observations point to inversion-symmetry breaking in the magnetic state. A weak Faraday feature near 320 cm$^{-1}$ ($\sim 40$ meV) further supports the presence of a higher-energy magnetic excitation. These findings underscore the intricate interplay between magnetic and lattice degrees of freedom in FePS$_3$ and establish its relevance for exploring spin-lattice interactions in two-dimensional antiferromagnetic materials.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 06:11:44 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 09:52:33 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Panda",
"Kartik",
""
],
[
"Kapon",
"Itzik",
""
],
[
"Dumcenco",
"Dumitru",
""
],
[
"van der Marel",
"Dirk",
""
],
[
"Kuzmenko",
"Alexey",
""
],
[
"Bachar",
"Nimrod",
""
]
] |
2507.17442
|
Shiting Chen
|
Shiting Chen, Zijian Zhao, Jinsong Chen
|
Each to Their Own: Exploring the Optimal Embedding in RAG
| null | null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Recently, as Large Language Models (LLMs) have fundamentally impacted various fields, the methods for incorporating up-to-date information into LLMs or adding external knowledge to construct domain-specific models have garnered wide attention. Retrieval-Augmented Generation (RAG), serving as an inference-time scaling method, is notable for its low cost and minimal effort for parameter tuning. However, due to heterogeneous training data and model architecture, the variant embedding models used in RAG exhibit different benefits across various areas, often leading to different similarity calculation results and, consequently, varying response quality from LLMs. To address this problem, we propose and examine two approaches to enhance RAG by combining the benefits of multiple embedding models, named Mixture-Embedding RAG and Confident RAG. Mixture-Embedding RAG simply sorts and selects retrievals from multiple embedding models based on standardized similarity; however, it does not outperform vanilla RAG. In contrast, Confident RAG generates responses multiple times using different embedding models and then selects the responses with the highest confidence level, demonstrating average improvements of approximately 10% and 5% over vanilla LLMs and RAG, respectively. The consistent results across different LLMs and embedding models indicate that Confident RAG is an efficient plug-and-play approach for various domains. We will release our code upon publication.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 12:03:54 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 06:44:38 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chen",
"Shiting",
""
],
[
"Zhao",
"Zijian",
""
],
[
"Chen",
"Jinsong",
""
]
] |
2507.17792
|
Jingyi Yu
|
Jingyi Yu, Tim Pychynski, and Marco F. Huber
|
Causal Mechanism Estimation in Multi-Sensor Systems Across Multiple Domains
|
To appear in 2025 28th International Conference on Information Fusion (FUSION)
| null | null | null |
cs.LG stat.ML
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
To gain deeper insights into a complex sensor system through the lens of causality, we present common and individual causal mechanism estimation (CICME), a novel three-step approach to inferring causal mechanisms from heterogeneous data collected across multiple domains. By leveraging the principle of Causal Transfer Learning (CTL), CICME is able to reliably detect domain-invariant causal mechanisms when provided with sufficient samples. The identified common causal mechanisms are further used to guide the estimation of the remaining causal mechanisms in each domain individually. The performance of CICME is evaluated on linear Gaussian models under scenarios inspired from a manufacturing process. Building upon existing continuous optimization-based causal discovery methods, we show that CICME leverages the benefits of applying causal discovery on the pooled data and repeatedly on data from individual domains, and it even outperforms both baseline methods under certain scenarios.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 10:35:37 GMT"
},
{
"version": "v2",
"created": "Fri, 25 Jul 2025 07:07:32 GMT"
},
{
"version": "v3",
"created": "Fri, 08 Aug 2025 07:28:53 GMT"
},
{
"version": "v4",
"created": "Wed, 20 Aug 2025 09:29:46 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Yu",
"Jingyi",
""
],
[
"Pychynski",
"Tim",
""
],
[
"Huber",
"Marco F.",
""
]
] |
2507.18272
|
Sandi Klav\v{z}ar
|
Csilla Bujt\'as and Vesna Ir\v{s}i\v{c} Chenoweth and Sandi Klav\v{z}ar and Gang Zhang
|
The $d$-distance $p$-packing domination number: complexity, cycles, and trees
| null | null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
A set of vertices $X\subseteq V(G)$ is a $d$-distance dominating set if for every $u\in V(G)\setminus X$ there exists $x\in X$ such that $d(u,x) \le d$, and $X$ is a $p$-packing if $d(u,v) \ge p+1$ for every different $u,v\in X$. The $d$-distance $p$-packing domination number $\gamma_d^p(G)$ of $G$ is the minimum size of a set of vertices of $G$ which is both a $d$-distance dominating set and a $p$-packing. It is proved that for every two fixed integers $d$ and $p$ with $2 \le d$ and $0 \le p \leq 2d-1$, the decision problem whether $\gamma_d^p(G) \leq k$ holds is NP-complete for bipartite planar graphs. A necessary and sufficient condition for the existence of a $d$-distance $p$-packing dominating set in $C_n$ is obtained and $\gamma_d^p(C_n)$ determined for every $d$, $p$, and $n$. For a tree $T$ on $n$ vertices with $\ell$ leaves and $s$ support vertices it is proved that (i) $\gamma_2^0(T) \geq \frac{n-\ell-s+4}{5}$, (ii) $\left \lceil \frac{n-\ell-s+4}{5} \right \rceil \leq \gamma_2^2(T) \leq \left \lfloor \frac{n+3s-1}{5} \right \rfloor$, and if $d \geq 2$, then (iii) $\gamma_d^2(T) \leq \frac{n-2\sqrt{n}+d+1}{d}$. Inequality (i) improves an earlier bound due to Meierling and Volkmann, and independently Raczek, Lema\'nska, and Cyman, while (iii) extends an earlier result for $\gamma_2^2(T)$ due to Henning. Sharpness of the bounds are discussed and established in most cases. It is also proved that every connected graph $G$ contains a spanning tree $T$ such that $\gamma_2^2(T) \leq \gamma_2^2(G)$.
|
[
{
"version": "v1",
"created": "Thu, 24 Jul 2025 10:17:49 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 09:26:24 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bujtás",
"Csilla",
""
],
[
"Chenoweth",
"Vesna Iršič",
""
],
[
"Klavžar",
"Sandi",
""
],
[
"Zhang",
"Gang",
""
]
] |
2507.19063
|
Alessandro Ruggieri
|
A. Ruggieri, A. Sozzetti, S. Desidera, D. Mesa, R. Gratton, F. Marzari, M. Bonavita, K. Biazzo, V. D'Orazi, C. Ginski, M. Meyer, L. Malavolta, M. Pinamonti, D. Barbato, C. Lazzoni, A. F. Lanza, L. Mancini, L. Naponiello, D. Nardiello, T. Zingales, M. Rainer, G. Scandariato, P. Giacobbe, R. Cosentino, A. Fiorenzano, R. Claudi
|
The GAPS Programme at TNG LXVIII. Characterization of the outer substellar companion around HD 72659 with a multi-technique approach
| null | null |
10.1093/mnras/staf1251
| null |
astro-ph.EP astro-ph.IM astro-ph.SR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Before discovering the first exoplanets, the Radial Velocity (RV) method had been used for decades to discover binary stars. Despite significant advancements in this technique, it is limited by the intrinsic mass-inclination degeneracy that can be broken when combining RVs with astrometry, which allows us to determine the orbital inclination, or direct imaging, from which we can estimate the true mass of the target. HD 72659 is a solar analog known to host a gas giant on a $\sim 10$-yr orbit and a massive outer companion. This work aims to confirm HD 72659 c, which was recently announced using data from HIRES and HARPS spectrographs in combination with Gaia's astrometric data. We monitored HD 72659 with HARPS-N in the framework of the GAPS project since 2012. We now combined our 91 spectra with literature data and Gaia DR3 high-precision astrometry to constrain the mass and the orbit of this object ($M_{\rm c} \sim 19$ $M_{\rm J}$, $a \sim 21$ au) that falls in the Brown Dwarf desert. Moreover, we analyzed our high-resolution imaging observation taken with SPHERE, but since the target was not detected, we could only derive upper limits on its mass. We characterize the orbital parameters of HD 72659 c, confirming the literature mass of this object but finding a period twice as high as previously reported, and we also refine the parameters of planet b with reduced uncertainties compared to previous works. Finally, we analyze and discuss the dynamic configuration of this system, finding that the Kozai-Lidov mechanism may be at work.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 08:25:17 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:57:42 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ruggieri",
"A.",
""
],
[
"Sozzetti",
"A.",
""
],
[
"Desidera",
"S.",
""
],
[
"Mesa",
"D.",
""
],
[
"Gratton",
"R.",
""
],
[
"Marzari",
"F.",
""
],
[
"Bonavita",
"M.",
""
],
[
"Biazzo",
"K.",
""
],
[
"D'Orazi",
"V.",
""
],
[
"Ginski",
"C.",
""
],
[
"Meyer",
"M.",
""
],
[
"Malavolta",
"L.",
""
],
[
"Pinamonti",
"M.",
""
],
[
"Barbato",
"D.",
""
],
[
"Lazzoni",
"C.",
""
],
[
"Lanza",
"A. F.",
""
],
[
"Mancini",
"L.",
""
],
[
"Naponiello",
"L.",
""
],
[
"Nardiello",
"D.",
""
],
[
"Zingales",
"T.",
""
],
[
"Rainer",
"M.",
""
],
[
"Scandariato",
"G.",
""
],
[
"Giacobbe",
"P.",
""
],
[
"Cosentino",
"R.",
""
],
[
"Fiorenzano",
"A.",
""
],
[
"Claudi",
"R.",
""
]
] |
2507.19157
|
Samuel Gaudout
|
Samuel Gaudout, Rayan Si-Ahmed, Cl\'ement Debavelaere, Menno Door, Pierre Clad\'e and Sa\"ida Guellati-Khelifa
|
Probing the spatial distribution of k-vectors in situ with Bose-Einstein condensates
| null | null | null | null |
physics.atom-ph quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a novel method for mapping in situ the spatial distribution of photon momentum across a laser beam using a Bose-Einstein condensate (BEC) as a moving probe. By displacing the BEC, we measure the photon recoil by atom interferometry at different positions in the laser beam and thus reconstruct a two-dimensional map of the local intensity and effective dispersion of the $k$ wave vector. Applied to a beam diffracted by a diaphragm, this method reveals a local extra recoil effect, which exceeds the magnitude $h\nu/c$ of the individual plane-waves over which the beam can be decomposed. This method offers a new way to precisely characterize wavefront distortions and to evaluate one of the major systematic bias sources in quantum sensors based on atom interferometry.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 10:58:58 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 13:04:51 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gaudout",
"Samuel",
""
],
[
"Si-Ahmed",
"Rayan",
""
],
[
"Debavelaere",
"Clément",
""
],
[
"Door",
"Menno",
""
],
[
"Cladé",
"Pierre",
""
],
[
"Guellati-Khelifa",
"Saïda",
""
]
] |
2507.21252
|
Himanshu Raj
|
Himanshu Raj and Raju Venugopalan
|
QCD-Gravity double copy in Regge asymptotics: from $2\rightarrow n$ amplitudes to radiation in shockwave collisions
|
136 pages, 40 figures, based on Lectures at the 64th Cracow School of Theoretical Physics, Zakopane, Poland; v2: typos corrected, references added
| null | null | null |
hep-th gr-qc hep-ph nucl-th
|
http://creativecommons.org/licenses/by/4.0/
|
These lectures discuss multi-particle production in QCD and in gravity at ultrarelativistic energies, their double copy relations, and strong parallels in emergent shockwave dynamics. Dispersive techniques are applied to derive the BFKL equation for multi-gluon production in Regge asymptotics. Identical methods apply in gravity and are captured by a gravitational Lipatov equation. The building blocks in both cases are Lipatov vertices and reggeized propagators satisfying double copy relations; in gravity, Weinberg's soft theorem is recovered as a limit of the Lipatov framework. BFKL evolution in QCD generates wee parton states of maximal occupancy characterized by an emergent semi-hard saturation scale. Renormalization group equations in the Color Glass Condensate (CGC) EFT describe wee parton correlations and their rapidity evolution. A shockwave picture of deeply inelastic scattering and hadron-hadron collisions follows, with multi-particle production described by Cutkosky's rules in strong time-dependent fields. Gluon radiation in the CGC EFT has a double copy in gravitational shockwave collisions, with a similar correspondence applicable between gluon and graviton shockwave propagators. Possible extensions of this semi-classical double copy are outlined for computing multi-particle production in gravitational shockwave collisions, self-force and tidal contributions, and classical and quantum noise in the focusing of geodesics.
|
[
{
"version": "v1",
"created": "Mon, 28 Jul 2025 18:15:29 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 21:39:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Raj",
"Himanshu",
""
],
[
"Venugopalan",
"Raju",
""
]
] |
2507.21733
|
Wolfgang Woess
|
Thomas Hirschler and Wolfgang Woess
|
The spectra of graph substitutions
| null | null | null | null |
math.CO math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
Let $(X,E_X)$ and $(V,E_V)$ be finite connected graphs without loops. We assume that $V$ has two distinguished vertices $a,b$ and an automorphism $\gamma$ which exchanges $a$ and~$b$. The $V$-edge substitution of $X$ is the graph $X[V]$ where each edge $[x,y] \in E_X$ is replaced by a copy of $V$, identifying $x$ with $a$ and $y$ with $b$ or vice versa. (The latter choice does not matter; it yields isomorphic graphs.) The aim is to describe the spectrum of $X[V]$ in terms of the spectra of $X$ and $V$. Instead of the spectra of the adjacency matrices, we consider the versions which are normalised by dividing each row by the row sum (the vertex degree). These are stochastic, reversible matrices, and our approach applies more generally to reversible transition matrices corresponding to arbitrary positive edge weights invariant under $\gamma$. We write $P$ for the transition matrix over $X$ and $Q$ for the one over $V$. Together, they induce the matrix $P_*$ over $X[V]$.
The main part of the spectrum of $P_*$ is the response of the natural frequencies of $X$ to substituting $V$, given by a functional equation coming from a rational function induced by~$Q$. A second part comes from specific eigenvalues of $Q$, if present. Finally, there is the part of $\mathsf{spec}(P_*)$ whose eigenfunctions have $X$ as a nodal set. The results depend on issues like whether $X$ has circles of even length and on the eigenvalues of the restriction of $Q$ to $V \setminus \{ a,b\}$, which are classified into 4 possible types. Quite subtle is the issue of determining the multiplicities of the latter as eigenvalues of $P_*$ in terms of the input.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 12:12:25 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 11:24:21 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hirschler",
"Thomas",
""
],
[
"Woess",
"Wolfgang",
""
]
] |
2507.21997
|
Andre Geim K
|
Daniil Domaretskiy, Zefei Wu, Van Huy Nguyen, Ned Hayward, Ian Babich, Xiao Li, Ekaterina Nguyen, Julien Barrier, Kornelia Indykiewicz, Wendong Wang, Roman V. Gorbachev, Na Xin, Kenji Watanabe, Takashi Taniguchi, Lee Hague, Vladimir I. Fal'ko, Irina V. Grigorieva, Leonid A. Ponomarenko, Alexey I. Berdyugin and Andre K. Geim
|
Proximity screening greatly enhances electronic quality of graphene
| null |
Nature 644, 646-651 (2025)
|
10.1038/s41586-025-09386-0
| null |
cond-mat.mes-hall cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The electronic quality of two-dimensional systems is crucial when exploring quantum transport phenomena. In semiconductor heterostructures, decades of optimization have yielded record-quality two-dimensional gases with transport and quantum mobilities reaching close to 10$^8$ and 10$^6$ cm$^2$/Vs, respectively. Although the quality of graphene devices has also been improving, it remains comparatively lower. Here we report a transformative improvement in the electronic quality of graphene by employing graphite gates placed in its immediate proximity, at 1 nm separation. The resulting screening reduces charge inhomogeneity by two orders of magnitude, bringing it down to a few 10$^7$ cm$^-2$ and limiting potential fluctuations to less than 1 meV. Quantum mobilities reach 10$^7$ cm$^2$/Vs, surpassing those in the highest-quality semiconductor heterostructures by an order of magnitude, and the transport mobilities match their record. This quality enables Shubnikov-de Haas oscillations in fields as low as 1 mT and quantum Hall plateaus below 5 mT. Although proximity screening predictably suppresses electron-electron interactions, fractional quantum Hall states remain observable with their energy gaps reduced only by a factor of 3-5 compared to unscreened devices, demonstrating that many-body phenomena at spatial scales shorter than 10 nm remain robust. Our results offer a reliable route to improving electronic quality in graphene and other two-dimensional systems, which should facilitate the exploration of new physics previously obscured by disorder.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 16:45:54 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Domaretskiy",
"Daniil",
""
],
[
"Wu",
"Zefei",
""
],
[
"Nguyen",
"Van Huy",
""
],
[
"Hayward",
"Ned",
""
],
[
"Babich",
"Ian",
""
],
[
"Li",
"Xiao",
""
],
[
"Nguyen",
"Ekaterina",
""
],
[
"Barrier",
"Julien",
""
],
[
"Indykiewicz",
"Kornelia",
""
],
[
"Wang",
"Wendong",
""
],
[
"Gorbachev",
"Roman V.",
""
],
[
"Xin",
"Na",
""
],
[
"Watanabe",
"Kenji",
""
],
[
"Taniguchi",
"Takashi",
""
],
[
"Hague",
"Lee",
""
],
[
"Fal'ko",
"Vladimir I.",
""
],
[
"Grigorieva",
"Irina V.",
""
],
[
"Ponomarenko",
"Leonid A.",
""
],
[
"Berdyugin",
"Alexey I.",
""
],
[
"Geim",
"Andre K.",
""
]
] |
2507.22417
|
Anik Chakraborty
|
Anik Chakraborty
|
Inverse Mobius Spacetime in 1+1D Quantum Gravity: Functional Analytic Structures, Dirac Spectrum, and Pin Geometry
|
12 pages, no figures. In this version, the author has revised the structure with subsections for clarity, added a conjecture on the Pin factor in the random matrix dual (Remark 5.5), included a new outlook paragraph on foundational gravitational questions with updated references (32 in total, including Einstein 1916 and Rovelli 2004), and made minor typographical fixes
| null | null | null |
gr-qc math-ph math.DG math.FA math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
In this manuscript, we formulate a (1+1)-dimensional Jackiw-Teitelboim gravity toy model whose Euclidean spacetime manifold is the M\"obius band $M$. Since $M$ is non-orientable, the relevant spin-statistics structure is Pin rather than Spin. To emphasize the role of orientation reversal, we refer to the universal orientable cover $\widetilde{M}$ as the inverse M\"obius band, which resolves the M\"obius twist into an infinite ribbon equipped with a $\mathbb{Z}$ deck action. We compute the Stiefel-Whitney classes $w_1$, $w_2$, classify all $\mathrm{Pin}^\pm$ structures, construct the associated pinor bundles, and analyze the Dirac operator under the twisted equivariance condition \[ \psi(x+1, w) = \gamma^w \psi(x, -w). \] Half-integer momentum quantization, spectral symmetry, vanishing mod-2 index, and $\eta_D(0) = 0$ follow. In JT gravity, the two inequivalent Pin lifts in each parity double the non-perturbative saddle-point sum, yet leave the leading Bekenstein-Hawking entropy unchanged. Full proofs and heat-kernel calculations are provided for completeness.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 06:42:42 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:51:25 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chakraborty",
"Anik",
""
]
] |
2507.22528
|
Dang Tuan Hiep
|
Dang Tuan Hiep and Khai-Hoan Nguyen-Dang
|
Supersymmetric Schur polynomials have saturated Newton polytopes
|
13 pages, 3 figures, add references, change abstract
| null | null | null |
math.CO math.RT
|
http://creativecommons.org/licenses/by/4.0/
|
We prove that every supersymmetric Schur polynomial has a saturated Newton polytope (SNP). Our approach begins with a tableau-theoretic description of the support, which we encode as a polyhedron with a totally unimodular constraint matrix. The integrality of this polyhedron follows from the Hoffman-Kruskal criterion, thereby establishing the SNP property.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 09:56:30 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 04:47:02 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hiep",
"Dang Tuan",
""
],
[
"Nguyen-Dang",
"Khai-Hoan",
""
]
] |
2507.23228
|
Wen Shen
|
Wen Shen
|
Manifold structures on highly connected Poincar\'e complexes
| null | null | null | null |
math.AT math.GT
|
http://creativecommons.org/publicdomain/zero/1.0/
|
This paper constructs numerous examples of highly connected Poincar\'{e} complexes, each homotopy equivalent to a topological manifold yet not homotopy equivalent to any smooth manifold. Furthermore, we determine the homotopy type of any closed $2k$-connected framed $(4k+2)$-manifold with Kervaire invariant one for $k=7,15,31$.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 03:50:06 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 07:56:09 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Shen",
"Wen",
""
]
] |
2507.23424
|
Josef Kluson
|
J. Kluson
|
Canonical Form of Born-Infeld Inspired Gravity Coupled to Scalar Fields
|
15 pages, references added
| null | null | null |
gr-qc
|
http://creativecommons.org/publicdomain/zero/1.0/
|
In this short note we find canonical form of Born-Infeld inspired gravity coupled to scalar fields using Faddeev-Jackiw approach. We show that canonical form of the action splits into two parts: First part has the same form as General Relativity canonical action while the matter part has complicated form as a consequence of the structure of Born-Infeld gravity action.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 11:03:07 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 18:40:33 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kluson",
"J.",
""
]
] |
2507.23477
|
Shenghao Hua
|
Shenghao Hua
|
Discrete restrictions from Laurent monomial systems for multiple Dirichlet series
|
6 pages. Comments welcome
| null | null | null |
math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce a special class of multiple Dirichlet series whose terms are supported on a variety and which admit an Euler product structure. We proposed several conjectures on the analytic properties of these series.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 11:57:15 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 18:11:08 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hua",
"Shenghao",
""
]
] |
2507.23479
|
Julia Werner
|
Julia Werner, Oliver Bause, Julius Oexle, Maxime Le Floch, Franz Brinkmann, Jochen Hampe and Oliver Bringmann
|
Seeing More with Less: Video Capsule Endoscopy with Multi-Task Learning
|
Accepted at Applications of Medical AI (AMAI workshop) at MICCAI 2025 (submitted version)
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Video capsule endoscopy has become increasingly important for investigating the small intestine within the gastrointestinal tract. However, a persistent challenge remains the short battery lifetime of such compact sensor edge devices. Integrating artificial intelligence can help overcome this limitation by enabling intelligent real-time decision-making, thereby reducing the energy consumption and prolonging the battery life. However, this remains challenging due to data sparsity and the limited resources of the device restricting the overall model size. In this work, we introduce a multi-task neural network that combines the functionalities of precise self-localization within the gastrointestinal tract with the ability to detect anomalies in the small intestine within a single model. Throughout the development process, we consistently restricted the total number of parameters to ensure the feasibility to deploy such model in a small capsule. We report the first multi-task results using the recently published Galar dataset, integrating established multi-task methods and Viterbi decoding for subsequent time-series analysis. This outperforms current single-task models and represents a significant advance in AI-based approaches in this field. Our model achieves an accuracy of 93.63% on the localization task and an accuracy of 87.48% on the anomaly detection task. The approach requires only 1 million parameters while surpassing the current baselines.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 12:00:25 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 14:59:44 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Werner",
"Julia",
""
],
[
"Bause",
"Oliver",
""
],
[
"Oexle",
"Julius",
""
],
[
"Floch",
"Maxime Le",
""
],
[
"Brinkmann",
"Franz",
""
],
[
"Hampe",
"Jochen",
""
],
[
"Bringmann",
"Oliver",
""
]
] |
2507.23707
|
Renato L. G. Cavalcante
|
Renato Luis Garrido Cavalcante and Tomasz Piotrowski and Slawomir Stanczak
|
Cellular, Cell-less, and Everything in Between: A Unified Framework for Utility Region Analysis in Wireless Networks
| null | null | null | null |
eess.SP cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce a unified framework for analyzing utility regions of wireless networks, with a focus on the signal-to-interference-noise-ratio (SINR) and achievable rate regions. The framework provides valuable insights into interference patterns of modern network architectures, such as cell-less and extremely large MIMO networks, and it generalizes existing characterizations of the weak Pareto boundary. A central contribution is the derivation of sufficient conditions that guarantee convexity of the utility regions. Convexity is an important property because it ensures that time sharing (or user grouping) cannot simultaneously increase the utility of all users when the network operates on the weak Pareto boundary. These sufficient conditions also have two key implications. First, they identify a family of (weighted) sum-rate maximization problems that are inherently convex without any variable transformations, thus paving the way for the development of efficient, provably optimal solvers for this family. Second, they provide a rigorous justification for formulating sum-rate maximization problems directly in terms of achievable rates, rather than SINR levels. Our theoretical insights also motivate an alternative to the concept of favorable propagation in the massive MIMO literature -- one that explicitly accounts for self-interference and the beamforming strategy.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 16:27:45 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 16:02:42 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Cavalcante",
"Renato Luis Garrido",
""
],
[
"Piotrowski",
"Tomasz",
""
],
[
"Stanczak",
"Slawomir",
""
]
] |
2508.00018
|
Ralph Marinaro
|
Ralph M. Marinaro III
|
Reconstructing Transition GPDs for Delta(1232) from Helicity Amplitude A_1/2(Q^2) via Dipole Fits and Impact Parameter Analysis
| null | null | null | null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This work presents a modular reconstruction of the transition generalized parton distribution (GPD) H_T(x,t) for the Delta(1232) resonance, based on digitized helicity amplitude data and dipole fits to A_1/2(Q^2). From the fitted amplitude, we extract a Sachs-like form factor F(t) and define a separable GPD model H_T(x, t) = h(x)F(t), with h(x) modeled as a normalized Beta-like profile. This factorized ansatz satisfies the GPD sum rule and enables a direct two-dimensional Fourier transform to construct transverse spatial distributions q(x,b). We analyze how longitudinal shaping modulates transverse localization, and quantify spatial features using statistical diagnostics including mean radius, skewness, and kurtosis. The framework is reproducible, data-driven, and applicable to other transition channels, providing a physically interpretable map from amplitude behavior to spatial structure.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 22:22:27 GMT"
},
{
"version": "v2",
"created": "Sat, 09 Aug 2025 02:59:52 GMT"
},
{
"version": "v3",
"created": "Wed, 20 Aug 2025 02:09:22 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Marinaro",
"Ralph M.",
"III"
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.