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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2507.17303
|
Zhe Xu
|
Zhe Xu, Ziyi Liu, Junlin Hou, Jiabo Ma, Cheng Jin, Yihui Wang, Zhixuan Chen, Zhengyu Zhang, Fuxiang Huang, Zhengrui Guo, Fengtao Zhou, Yingxue Xu, Xi Wang, Ronald Cheong Kin Chan, Li Liang, Hao Chen
|
A Versatile Pathology Co-pilot via Reasoning Enhanced Multimodal Large Language Model
| null | null | null | null |
eess.IV cs.AI cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Multimodal large language models (MLLMs) have emerged as powerful tools for computational pathology, offering unprecedented opportunities to integrate pathological images with language context for comprehensive diagnostic analysis. These models hold particular promise for automating complex tasks that traditionally require expert interpretation of pathologists. However, current MLLM approaches in pathology demonstrate significantly constrained reasoning capabilities, primarily due to their reliance on expensive chain-of-thought annotations. Additionally, existing methods remain limited to simplex application of visual question answering (VQA) at the region-of-interest (ROI) level, failing to address the full spectrum of diagnostic needs such as ROI classification, detection, segmentation, whole-slide-image (WSI) classification and VQA in clinical practice. In this study, we present SmartPath-R1, a versatile MLLM capable of simultaneously addressing both ROI-level and WSI-level tasks while demonstrating robust pathological reasoning capability. Our framework combines scale-dependent supervised fine-tuning and task-aware reinforcement fine-tuning, which circumvents the requirement for chain-of-thought supervision by leveraging the intrinsic knowledge within MLLM. Furthermore, SmartPath-R1 integrates multiscale and multitask analysis through a mixture-of-experts mechanism, enabling dynamic processing for diverse tasks. We curate a large-scale dataset comprising 2.3M ROI samples and 188K WSI samples for training and evaluation. Extensive experiments across 72 tasks validate the effectiveness and superiority of the proposed approach. This work represents a significant step toward developing versatile, reasoning-enhanced AI systems for precision pathology.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 08:09:42 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:51:01 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Xu",
"Zhe",
""
],
[
"Liu",
"Ziyi",
""
],
[
"Hou",
"Junlin",
""
],
[
"Ma",
"Jiabo",
""
],
[
"Jin",
"Cheng",
""
],
[
"Wang",
"Yihui",
""
],
[
"Chen",
"Zhixuan",
""
],
[
"Zhang",
"Zhengyu",
""
],
[
"Huang",
"Fuxiang",
""
],
[
"Guo",
"Zhengrui",
""
],
[
"Zhou",
"Fengtao",
""
],
[
"Xu",
"Yingxue",
""
],
[
"Wang",
"Xi",
""
],
[
"Chan",
"Ronald Cheong Kin",
""
],
[
"Liang",
"Li",
""
],
[
"Chen",
"Hao",
""
]
] |
2507.17549
|
Anton Izosimov
|
Anton Izosimov
|
Folding of quadrilaterals, zigzags, and Arnold-Liouville integrability
|
5 pages, 7 figures, accepted to Arnold Mathematical Journal
| null | null |
MPIM-Bonn-2025
|
nlin.SI math.MG math.SG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We put Darboux's porism on folding of quadrilaterals, as well as closely related Bottema's zigzag porism, in the context of Arnold-Liouville integrability.
|
[
{
"version": "v1",
"created": "Wed, 23 Jul 2025 14:28:33 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Izosimov",
"Anton",
""
]
] |
2507.18752
|
Kevork Abazajian
|
Cannon M. Vogel, Helena Garc\'ia Escudero, Julien Froustey, Kevork N. Abazajian
|
Return of the Lepton Number: Sterile Neutrino Dark Matter Production and the Revival of the Shi-Fuller Mechanism
|
13 pages, 6 figures; v2: added references and updated plots
| null | null |
UCI-TR-2025-08, N3AS-25-012
|
hep-ph astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We explore resonant production of sterile neutrino dark matter via the Shi-Fuller (SF) mechanism, revisiting its cosmological viability in light of recent results demonstrating that lepton-number asymmetries $L_\alpha \gtrsim 1$ at temperatures $T > 20\rm\,MeV$ are consistent with Big Bang Nucleosynthesis (BBN). Using a quasi-classical Boltzmann transport calculation of the dark matter production, we compute the non-thermal phase space distributions of sterile neutrinos across a broad range of particle mass $m_s$ and mixing angle $\sin^2{(2\theta)}$ parameter space. We then evolve the resulting distributions through linear structure formation using CLASS and fit the resulting matter power spectra to thermal warm dark matter (WDM) transfer functions, enabling a direct mapping between SF models and equivalent thermal WDM particle masses $m_{\mathrm{th}}$. This allows us to reinterpret existing structure formation limits and Lyman-$\alpha$ forest preferences in the context of SF production. We find that lepton asymmetries $L \gtrsim 0.5$ at high temperatures open significant viable parameter space in the $m_s \gtrsim 10\,\mathrm{keV}$ and $\sin^2 (2\theta) \lesssim 10^{-14}$ regime, compatible with both X-ray constraints from NuSTAR and INTEGRAL/SPI and recent Lyman-$\alpha$ inferences of $m_{\mathrm{th}} \approx 4.1\,\mathrm{keV}$. Following lepton number evolution below 20 MeV, we also specifically show that this lepton asymmetry parameter space is compatible with BBN and cosmic microwave background constraints. We present updated constraints, a refined $m_{\mathrm{th}}$ fitting function, and power-law approximations for $L$ across the parameter space. Our results motivate future X-ray observations targeting the $\sim 20$ keV photon regime and testing of the $m_\mathrm{th} \gtrsim 10\,\mathrm{keV}$ WDM region.
|
[
{
"version": "v1",
"created": "Thu, 24 Jul 2025 19:03:41 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 22:50:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Vogel",
"Cannon M.",
""
],
[
"Escudero",
"Helena García",
""
],
[
"Froustey",
"Julien",
""
],
[
"Abazajian",
"Kevork N.",
""
]
] |
2507.19046
|
Charlotte Geier
|
Charlotte Geier (1), Rasha Shanaz (2), and Merten Stender (3) ((1) Dynamics Group, Hamburg University of Technology, Germany, (2) Department of Physics, Bharathidasan University, Tiruchirappalli, India, (3) Chair of Cyber-Physical Systems in Mechanical Engineering, Technische Universit\"at Berlin, Germany)
|
Dynamics-Informed Reservoir Computing with Visibility Graphs
|
7 pages, 5 figures. The following article has been submitted to by Chaos: An Interdisciplinary Journal of Nonlinear Science
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Accurate prediction of complex and nonlinear time series remains a challenging problem across engineering and scientific disciplines. Reservoir computing (RC) offers a computationally efficient alternative to traditional deep learning by training only the read-out layer while employing a randomly structured and fixed reservoir network. Despite its advantages, the largely random reservoir graph architecture often results in suboptimal and oversized networks with poorly understood dynamics. Addressing this issue, we propose a novel Dynamics-Informed Reservoir Computing (DyRC) framework that systematically infers the reservoir network structure directly from the input training sequence. This work proposes to employ the visibility graph (VG) technique, which converts time series data into networks by representing measurement points as nodes linked by mutual visibility. The reservoir network is constructed by directly adopting the VG network from a training data sequence, leveraging the parameter-free visibility graph approach to avoid expensive hyperparameter tuning. This process results in a reservoir that is directly informed by the specific dynamics of the prediction task under study. We assess the DyRC-VG method through prediction tasks involving the canonical nonlinear Duffing oscillator, evaluating prediction accuracy and consistency. Compared to an Erd\H{o}s-R\'enyi (ER) graph of the same size, spectral radius, and fixed density, we observe higher prediction quality and more consistent performance over repeated implementations in the DyRC-VG. An ER graph with density matched to the DyRC-VG can in some conditions outperform both approaches.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 08:07:17 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:25:34 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Geier",
"Charlotte",
"",
"Dynamics Group, Hamburg University of Technology, Germany"
],
[
"Shanaz",
"Rasha",
"",
"Department of Physics, Bharathidasan University, Tiruchirappalli, India"
],
[
"Stender",
"Merten",
"",
"Chair of Cyber-Physical Systems in Mechanical Engineering, Technische Universität Berlin, Germany"
]
] |
2507.19094
|
Nolwenn Maudet
|
Thomas Thibault, L\'ea Mosesso, Camille Adam, Aur\'elien Tabard, Ana\"elle Beignon and Nolwenn Maudet
|
Environmental (in)considerations in the Design of Smartphone Settings
|
Post-proceedings paper presented at LIMITS 2025: 11th Workshop on Computing within Limits, 2025-06-26/27, Online
| null | null | null |
cs.HC
|
http://creativecommons.org/licenses/by/4.0/
|
Designing for sufficiency is one of many approaches that could foster more moderate and sustainable digital practices. Based on the Sustainable Information and Communication Technologies (ICT) and Human-Computer Interaction (HCI) literature, we identify five environmental settings categories. However, our analysis of three mobile OS and nine representative applications shows an overall lack of environmental concerns in settings design, leading us to identify six pervasive anti-patterns. Environmental settings, where they exist, are set on the most intensive option by default. They are not presented as such, are not easily accessible, and offer little explanation of their impact. Instead, they encourage more intensive use. Based on these findings, we create a design workbook that explores design principles for environmental settings: presenting the environmental potential of settings; shifting to environmentally neutral states; previewing effects to encourage moderate use; rethinking defaults; facilitating settings access and; exploring more frugal settings. Building upon this workbook, we discuss how settings can tie individual behaviors to systemic factors.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 09:24:26 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:46:08 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Thibault",
"Thomas",
""
],
[
"Mosesso",
"Léa",
""
],
[
"Adam",
"Camille",
""
],
[
"Tabard",
"Aurélien",
""
],
[
"Beignon",
"Anaëlle",
""
],
[
"Maudet",
"Nolwenn",
""
]
] |
2507.19263
|
Achille Morenville
|
Achille Morenville and \'Eric Piette
|
Modeling Uncertainty: Constraint-Based Belief States in Imperfect-Information Games
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
In imperfect-information games, agents must make decisions based on partial knowledge of the game state. The Belief Stochastic Game model addresses this challenge by delegating state estimation to the game model itself. This allows agents to operate on externally provided belief states, thereby reducing the need for game-specific inference logic. This paper investigates two approaches to represent beliefs in games with hidden piece identities: a constraint-based model using Constraint Satisfaction Problems and a probabilistic extension using Belief Propagation to estimate marginal probabilities. We evaluated the impact of both representations using general-purpose agents across two different games. Our findings indicate that constraint-based beliefs yield results comparable to those of probabilistic inference, with minimal differences in agent performance. This suggests that constraint-based belief states alone may suffice for effective decision-making in many settings.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 13:38:44 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 12:30:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Morenville",
"Achille",
""
],
[
"Piette",
"Éric",
""
]
] |
2507.19580
|
Zheng Zhou
|
Zheng Zhou, Chong Wang, and Yin-Chen He
|
Chern-Simons-matter conformal field theory on fuzzy sphere: Confinement transition of Kalmeyer-Laughlin chiral spin liquid
|
15+12 Pages, 4+3 figures, 0+4 tables
| null | null | null |
cond-mat.str-el cond-mat.mes-hall cond-mat.stat-mech hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Gauge theories compose a large class of interacting conformal field theories in 3d, among which an outstanding category is critical Chern-Simons-matter theories. In this paper, we focus on one of the simplest instances: one complex critical scalar coupled to $\mathrm{U}(1)_2$ Chern-Simons gauge field. It is theoretically interesting as it is conjectured to exhibit dualities between four simple Lagrangian descriptions, but also practically important as it describes the transition between Kalmeyer-Laughlin chiral spin liquid (or $\nu=1/2$ bosonic Laughlin state) and trivially gapped phase. Using the fuzzy sphere regularisation, we realise this theory as a transition on the spherical lowest Landau level between a $\nu_f=2$ fermionic integer quantum Hall state and a $\nu_b=1/2$ bosonic fractional quantum Hall state. We show that this transition is continuous and has emergent conformal symmetry. By studying the operator spectrum, we show that there exists only one relevant singlet with scaling dimension $\Delta_S=1.52(18)$. We also discuss other higher operators and the consequences of our results.
|
[
{
"version": "v1",
"created": "Fri, 25 Jul 2025 18:00:00 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 17:57:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhou",
"Zheng",
""
],
[
"Wang",
"Chong",
""
],
[
"He",
"Yin-Chen",
""
]
] |
2507.20214
|
Nazl{\i} Do\u{g}an
|
Nazl{\i} Do\u{g}an
|
The Rhaly Operators on K\"othe Spaces
|
The definition of power boundedness has been updated to include a constant , and the related results have been revised accordingly
| null | null | null |
math.FA math.CV
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We introduce and study the Rhaly operator on K\"othe spaces, with a primary focus on understanding its well-definedness, continuity, and compactness. We especially examine operators acting on power series spaces of both infinite and finite type. In the sequel, we provide integral representations for the Rhaly operator on the space of entire functions $H(\mathbb{C})$ and the space of holomorphic functions on the unit disc $H(\mathbb{D})$. We also investigate the topologizability and power boundedness of the Rhaly operators, which leads to findings about their mean ergodicity, uniform mean ergodicity, and Ces\`aro boundedness.
|
[
{
"version": "v1",
"created": "Sun, 27 Jul 2025 10:40:25 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 19:19:07 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Doğan",
"Nazlı",
""
]
] |
2507.20845
|
Erica Bertolini
|
Erica Bertolini, Edoardo Lui and Nicola Maggiore
|
Quasi-topological mass generation for 3D linearized gravity
|
22 pages, no figures, accepted for publication in Phys. Rev. D
|
Phys. Rev. D 112, 044035 (2025)
|
10.1103/db71-jwd2
|
DIAS-STP-25-18
|
hep-th gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
We present a new mass generation mechanism for linearized gravity in three spacetime dimensions, which consists of a lower-dimensional Chern-Simons-like term added to the invariant action. The propagators of the gauge fixed massive action show a massive pole and a good massless limit. Moreover, we show that, as the Topological Massive Gravity model of Deser, Jackiw and Templeton, this theory displays one propagating massive DoF, which can be traced back to the transverse part of the spatial Ricci tensor. Finally, the action of this linearized massive gravity is characterized by an algebraic structure formed by a set of Ward operators, which uniquely determine the theory.
|
[
{
"version": "v1",
"created": "Mon, 28 Jul 2025 13:54:30 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bertolini",
"Erica",
""
],
[
"Lui",
"Edoardo",
""
],
[
"Maggiore",
"Nicola",
""
]
] |
2507.20891
|
Matias Mazzanti
|
Matias Mazzanti, Augusto Vega and Esteban Mocskos
|
Characterizing the Sensitivity to Individual Bit Flips in Client-Side Operations of the CKKS Scheme
| null | null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Homomorphic Encryption (HE) enables computation on encrypted data without decryption, making it a cornerstone of privacy-preserving computation in untrusted environments. As HE sees growing adoption in sensitive applications such as secure machine learning and confidential data analysis ensuring its robustness against errors becomes critical. Faults (e.g., transmission errors, hardware malfunctions, or synchronization failures) can corrupt encrypted data and compromise the integrity of HE operations. However, the impact of soft errors (such as bit flips) on modern HE schemes remains unexplored. Specifically, the CKKS scheme-one of the most widely used HE schemes for approximate arithmetic-lacks a systematic study of how such errors propagate across its pipeline, particularly under optimizations like the Residue Number System (RNS) and Number Theoretic Transform (NTT). This work bridges that gap by presenting a theoretical and empirical analysis of CKKS's fault tolerance under single bit-flip errors. We focus on client-side operations (encoding, encryption, decryption, and decoding) and demonstrate that while the vanilla CKKS scheme exhibits some resilience, performance optimizations (RNS/NTT) introduce significant fragility, amplifying error sensitivity. By characterizing these failure modes, we lay the groundwork for error-resilient HE designs, ensuring both performance and integrity in privacy-critical applications.
|
[
{
"version": "v1",
"created": "Mon, 28 Jul 2025 14:42:09 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 18:55:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Mazzanti",
"Matias",
""
],
[
"Vega",
"Augusto",
""
],
[
"Mocskos",
"Esteban",
""
]
] |
2507.21447
|
Sunita Chandrasekaran
|
Zachariah Sollenberger, Rahul Patel, Saieda Ali Zada, Sunita Chandrasekaran
|
LLM4VV: Evaluating Cutting-Edge LLMs for Generation and Evaluation of Directive-Based Parallel Programming Model Compiler Tests
| null | null | null | null |
cs.SE cs.ET
|
http://creativecommons.org/licenses/by/4.0/
|
The usage of Large Language Models (LLMs) for software and test development has continued to increase since LLMs were first introduced, but only recently have the expectations of LLMs become more realistic. Verifying the correctness of code generated by LLMs is key to improving their usefulness, but there have been no comprehensive and fully autonomous solutions developed yet. Hallucinations are a major concern when LLMs are applied blindly to problems without taking the time and effort to verify their outputs, and an inability to explain the logical reasoning of LLMs leads to issues with trusting their results. To address these challenges while also aiming to effectively apply LLMs, this paper proposes a dual-LLM system (i.e. a generative LLM and a discriminative LLM) and experiments with the usage of LLMs for the generation of a large volume of compiler tests. We experimented with a number of LLMs possessing varying parameter counts and presented results using ten carefully-chosen metrics that we describe in detail in our narrative. Through our findings, it is evident that LLMs possess the promising potential to generate quality compiler tests and verify them automatically.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 02:34:28 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 03:08:54 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Sollenberger",
"Zachariah",
""
],
[
"Patel",
"Rahul",
""
],
[
"Zada",
"Saieda Ali",
""
],
[
"Chandrasekaran",
"Sunita",
""
]
] |
2507.21777
|
Xiao-Xiong Zeng
|
Xiao-Xiong Zeng, Ke Wang
|
Energy Extraction From the Kerr-Bertotti-Robinson Black Hole via Magnetic Reconnection under Circular Plasma and Plunging Plasma
|
20 pages, 10 figures, final version
| null | null | null |
gr-qc hep-th
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Recently, a class of exact solutions describing rotating black holes immersed in a uniform magnetic field has been proposed, and various properties of such black holes remain unclear. This paper aims to explore how to extract energy based on the magnetic reconnection mechanism in both the circular orbit region and the plunging region. After introducing the properties of this spacetime, we analyze physical quantities such as the size of the ergoregion, the event horizon, and the boundaries of the ergosphere. We then analyze the magnetic reconnection process within circular orbits. We plot energy extraction parameter diagrams, and analyze the power and efficiency of energy extraction. The results show that the magnetic field impedes energy extraction. Comparing with Kerr black holes and Kerr-Melvin black holes, we find that this black hole's energy extraction capability is stronger than that of the Kerr-Melvin black hole but weaker than that of the Kerr black hole. Furthermore, we study magnetic reconnection and energy extraction in the plunging region of this black hole. The results indicate that the magnetic field still impedes energy extraction. The black hole's energy extraction capability in this region also remains weaker than that of a Kerr black hole but stronger than that of a Kerr-Melvin black hole. Comparing the plunging region with the circular orbit region, we find that the energy extraction power in the plunging region is consistently stronger than in the circular orbit region.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 13:01:44 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 07:13:27 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zeng",
"Xiao-Xiong",
""
],
[
"Wang",
"Ke",
""
]
] |
2507.21830
|
Fanda Fan
|
Kuiye Ding, Fanda Fan, Yao Wang, Ruijie jian, Xiaorui Wang, Luqi Gong, Yishan Jiang, Chunjie Luo, Jianfeng Zhan
|
DualSG: A Dual-Stream Explicit Semantic-Guided Multivariate Time Series Forecasting Framework
|
This paper has been accepted by ACM Multimedia 2025 (ACM MM 2025)
| null |
10.1145/3746027.3755458
| null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Multivariate Time Series Forecasting plays a key role in many applications. Recent works have explored using Large Language Models for MTSF to take advantage of their reasoning abilities. However, many methods treat LLMs as end-to-end forecasters, which often leads to a loss of numerical precision and forces LLMs to handle patterns beyond their intended design. Alternatively, methods that attempt to align textual and time series modalities within latent space frequently encounter alignment difficulty. In this paper, we propose to treat LLMs not as standalone forecasters, but as semantic guidance modules within a dual-stream framework. We propose DualSG, a dual-stream framework that provides explicit semantic guidance, where LLMs act as Semantic Guides to refine rather than replace traditional predictions. As part of DualSG, we introduce Time Series Caption, an explicit prompt format that summarizes trend patterns in natural language and provides interpretable context for LLMs, rather than relying on implicit alignment between text and time series in the latent space. We also design a caption-guided fusion module that explicitly models inter-variable relationships while reducing noise and computation. Experiments on real-world datasets from diverse domains show that DualSG consistently outperforms 15 state-of-the-art baselines, demonstrating the value of explicitly combining numerical forecasting with semantic guidance.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 14:08:09 GMT"
},
{
"version": "v2",
"created": "Wed, 30 Jul 2025 07:57:31 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 02:47:49 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ding",
"Kuiye",
""
],
[
"Fan",
"Fanda",
""
],
[
"Wang",
"Yao",
""
],
[
"jian",
"Ruijie",
""
],
[
"Wang",
"Xiaorui",
""
],
[
"Gong",
"Luqi",
""
],
[
"Jiang",
"Yishan",
""
],
[
"Luo",
"Chunjie",
""
],
[
"Zhan",
"Jianfeng",
""
]
] |
2507.21951
|
Shenghao Hua
|
Shenghao Hua
|
Quadratic forms of holomorphic cusp forms and the decay of their $\ell^p$-norms for $0 < p < 2$
|
14 pages. A small change in the statement of the main theorem based on the referees' helpful comments
| null | null | null |
math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we demonstrate that, given an orthonormal basis of holomorphic Hecke cusp forms, conditionally, quadratic forms composed of cusp forms -- each expressed as a bounded linear combination of holomorphic Hecke cusp forms -- are generally not themselves expressible as bounded linear combinations of holomorphic Hecke cusp forms when the sum of the weights exceeds some absolute constant, provided that the coefficients of the quadratic form satisfy appropriate nonvanishing and boundedness conditions. This illustrates the finiteness of the number of solutions to the linear equation of modular forms equated to a quadratic form of large weight.
We also show that, conditionally, for $0 < p < 2$, the $\ell^p$-norm of such quadratic forms in holomorphic Hecke cusp forms tends to zero asymptotically with respect to expansion in this orthonormal basis of Hecke eigenforms.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 16:02:43 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:29:44 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hua",
"Shenghao",
""
]
] |
2507.22179
|
Jacob Spertus
|
Jacob V Spertus and Amanda K Glazer and Philip B Stark
|
Dice, but don't slice: Optimizing the efficiency of ONEAudit
|
16 pages, 3 tables, accepted to E-Vote ID 2025
| null | null | null |
stat.ME
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
ONEAudit provides more efficient risk-limiting audits than other extant methods when the voting system cannot report a cast-vote record linked to each cast card. It obviates the need for re-scanning; it is simpler and more efficient than 'hybrid' audits; and it is far more efficient than batch-level comparison audits. There may be room to improve the efficiency of ONEAudit further by tuning the statistical tests it uses and by using stratified sampling. We show that tuning the tests by optimizing for the reported batch-level tallies or integrating over a distribution reduces expected workloads by 70-85% compared to the current ONEAudit implementation across a range of simulated elections. The improved tests reduce the expected workload to audit the 2024 Mayoral race in San Francisco, California, by half -- from about 200 cards to about 100 cards. In contrast, stratified sampling does not help: it increases workloads by about 25% on average.
|
[
{
"version": "v1",
"created": "Tue, 29 Jul 2025 19:17:09 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 18:39:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Spertus",
"Jacob V",
""
],
[
"Glazer",
"Amanda K",
""
],
[
"Stark",
"Philip B",
""
]
] |
2507.22320
|
Shuto Tago
|
Shuto Tago, Akiyoshi Yamada, Yuki Fuseya
|
Magnetoresistance in the Extreme Quantum Limit: Field-Induced Crossover to the Unitarity Limit
| null |
2025 J. Phys.: Condens. Matter 37 325701
|
10.1088/1361-648X/adf538
|
32
|
cond-mat.mtrl-sci
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We theoretically investigate magnetoresistance (MR) in the extreme quantum limit (EQL), where the kinetic energy becomes significantly smaller than the cyclotron energy, using the Kubo formula with Green's functions and the $T$-matrix approximation. We uncover a magnetic-field-induced crossover in the scattering rate: $1/\tau \propto B^2$ in the Born regime and $1/\tau \propto B^{-2}$ in the unitarity limit. This crossover gives rise to distinct MR behaviors in the EQL, characterized by linear transverse MR ($\rho_{xx} \propto B$) and negative longitudinal MR ($\rho_{zz} \propto B^{-2}$). This dichotomy implies insulating behavior when the magnetic field is perpendicular to the current, and metallic behavior when it is parallel. In the unitarity limit, we further derive a universal relation that enables direct experimental determination of the impurity density from $\rho_{xx}$ and $\rho_{xy}$. Our results establish a quantum--classical correspondence that remains valid even in the EQL, provided that the field dependences of the scattering rate and quantum corrections are properly incorporated.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 01:28:56 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Tago",
"Shuto",
""
],
[
"Yamada",
"Akiyoshi",
""
],
[
"Fuseya",
"Yuki",
""
]
] |
2507.22595
|
Daniel Vainshtein
|
Amit Giveon and Daniel Vainshtein
|
On String theory on deformed BTZ and $T\bar{T} + J\bar{T} + T\bar{J}$
|
19 Pages
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
Aspects of superstring theory on black-strings backgrounds, corresponding to deformed BTZ black holes, formed near $k$ NS5 branes by $p$ fundamental strings, and single-trace $T\bar T+J\bar T+T\bar J$ holography, are presented. It is shown that for a particular asymptotic value of the $B$-field, the excitation energy of a long string plus its contribution to the energy of the black hole is that in $T\bar T+J\bar T+T\bar J$ deformed $CFT_2$ with $c=6k$. The excitation energy of a winding $w>1$ long string plus its contribution to the background, a $w/p$ fraction of the black-hole energy, evolves according to that in a $Z_w$ twisted sector of a $p$-fold symmetric product of the latter.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 11:49:51 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 10:00:00 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Giveon",
"Amit",
""
],
[
"Vainshtein",
"Daniel",
""
]
] |
2507.22838
|
Tan Tran
|
Tan Tran, Denisa Martonova, Sigrid Leyendecker
|
Modelling and simulation of electro-mechanically coupled dielectric elastomers and myocardial tissue using smoothed finite element methods
| null | null | null | null |
cs.CE
|
http://creativecommons.org/licenses/by/4.0/
|
Computational modelling offers a cost-effective and time-efficient alternative to experimental studies in biomedical engineering. In cardiac electro-mechanics, finite element method (FEM)-based simulations provide valuable insights into diseased tissue behaviour and the development of assistive systems such as di-electric elastomer actuators. However, the use of automatically generated tetrahedral meshes, commonly applied due to geometric complexity, often leads to numerical issues including overly stiff responses and volume locking, particularly in incompressible materials. Smoothed finite element methods (S-FEMs) offer a promising alternative by softening the stiffness matrix through gradient smoothing over defined smoothing domains. This work extends S-FEM formulations to electro-mechanically coupled problems and compares their performance against standard linear FEM. We implement and evaluate four approaches in the Abaqus environment via custom user elements: standard linear FEM, face-based S-FEM (FS-FEM), node-based S-FEM (NS-FEM), and the hybrid face/node-based S-FEM (FSNS-FEM). Two benchmark problems are studied: the electrically induced contraction of a compressible dielectric elastomer and an incompressible, orthotropic myocardial tissue sample. Reference solutions are obtained using a mesh consisting of higher-order elements. Our results demonstrate that FSNS-FEM provides the best balance between accuracy and computational efficiency, closely matching reference data. NS-FEM produces softer results, which leads to an overestimation of the true deformation. FS-FEM and standard FEM consistently exhibit overly stiff behaviour, with pronounced volume locking in the myocardial case. These findings support the potential of S-FEMs, in particular FSNS-FEM, for accurate simulation of coupled electro-mechanical behaviour in complex biomedical applications.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 16:53:49 GMT"
},
{
"version": "v2",
"created": "Thu, 31 Jul 2025 11:01:43 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 10:56:36 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Tran",
"Tan",
""
],
[
"Martonova",
"Denisa",
""
],
[
"Leyendecker",
"Sigrid",
""
]
] |
2507.23099
|
Kareem Elgindy
|
Kareem T. Elgindy
|
Hybrid Shifted Gegenbauer Integral-Pseudospectral Method for Solving Time-Fractional Benjamin-Bona-Mahony-Burgers Equation
| null | null | null | null |
math.NA cs.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper presents a high-order hybrid shifted Gegenbauer integral-pseudospectral (HSG-IPS) method for solving the time-fractional Benjamin-Bona-Mahony-Burgers (FBBMB) equation. A key innovation of our approach is the transformation of the original equation into a fractional partial-integro differential form that contains only a first-order derivative, which can be accurately approximated using a first-order shifted Gegenbauer differentiation matrix (SGDM), while all other terms in the transformed equation are resolved using highly accurate quadrature rules. The method combines several advanced numerical techniques including the shifted Gegenbauer pseudospectral (SGPS) method, Gegenbauer-based fractional approximation (GBFA), shifted Gegenbauer integration matrix (SGIM), shifted Gegenbauer integration row vector (SGIRV), and SGDM to achieve spectral accuracy. Numerical experiments demonstrate that the HSG-IPS method outperforms existing numerical approaches, achieving significantly lower average absolute errors (AAEs) with computational times as low as 0.04-0.05 seconds. The method's robustness is validated across various fractional orders, showing excellent agreement with analytical solutions. The transformation strategy effectively circumvents the numerical instability associated with direct approximation of high-order derivatives in the original equation, while the use of shifted Gegenbauer (SG) polynomials and barycentric representations ensures numerical stability and efficiency. This work provides a powerful computational framework for modeling wave propagation, dispersion, and nonlinearity in fractional calculus applications.
|
[
{
"version": "v1",
"created": "Wed, 30 Jul 2025 21:00:30 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:50:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Elgindy",
"Kareem T.",
""
]
] |
2507.23345
|
Takashi Ishizuka
|
Takashi Ishizuka
|
The PPP-completeness of the Ward-Szabo theorem
| null | null | null | null |
cs.CC cs.DM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Ward and Szab\'o [WS94] have shown that a complete graph with $N^2$ nodes whose edges are colored by $N$ colors and that has at least two colors contains a bichromatic triangle. This fact leads us to a total search problem: Given an edge-coloring on a complete graph with $N^2$ nodes using at least two colors and at most $N$ colors, find a bichromatic triangle. Bourneuf, Folwarczn\'y, Hub\'acek, Rosen, and Schwartzbach [Bou+23] have proven that such a total search problem, called Ward-Szab\'o, is PWPP-hard and belongs to the class TFNP, a class for total search problems in which the correctness of every candidate solution is efficiently verifiable. However, it is open which TFNP subclass contains Ward-Szab\'o. This paper will improve the computational complexity of Ward-Szab\'o. We prove that Ward-Szab\'o is a complete problem for the complexity class PPP, a TFNP subclass of problems in which the existence of solutions is guaranteed by the pigeonhole principle.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 08:50:40 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 04:07:11 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ishizuka",
"Takashi",
""
]
] |
2507.23387
|
Weicheng Xue
|
Weicheng Xue, Baisong Xu, Kai Yang, Yongxiang Liu, Dengdeng Fan, Pengxiang Xu, Yonghong Tian
|
SGEMM-cube: Emulating FP32 GEMM on Ascend NPUs Using FP16 Cube Units with Precision Recovery
| null | null | null | null |
cs.DC
|
http://creativecommons.org/licenses/by/4.0/
|
Low-precision matrix engines, such as FP16 cube, offer high throughput but lack support for full-precision computation. In this work, we propose SGEMM-cube, a high-performance algorithm for emulating FP32 general matrix-matrix multiplication (GEMM) using only FP16 computation units on a representative AI accelerator. The method decomposes each FP32 operand into two FP16 values and compensates for numerical errors through a tunable scaling strategy. A detailed analysis of numerical errors, including underflow conditions and precision loss, guides the selection of scaling parameters to preserve up to 22 bits of mantissa accuracy. We further investigate the effect of computation order on accuracy and demonstrate that a term-wise accumulation scheme improves numerical stability over conventional FP32 GEMM in low-exponent regimes. Finally, a cache-aware blocking strategy and double-buffered pipeline are introduced to overlap memory transfers with computation, enabling SGEMM-cube to achieve up to 77\% of the theoretical FP32-equivalent peak performance on Ascend 910A NPU lacking native FP32 support. Extensive numerical experiments confirm that our method not only recovers the accuracy of native FP32 GEMM but also exhibits superior numerical stability under certain conditions, due to its structured and error-aware computation order.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 10:02:26 GMT"
},
{
"version": "v2",
"created": "Fri, 01 Aug 2025 03:43:24 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 09:13:13 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Xue",
"Weicheng",
""
],
[
"Xu",
"Baisong",
""
],
[
"Yang",
"Kai",
""
],
[
"Liu",
"Yongxiang",
""
],
[
"Fan",
"Dengdeng",
""
],
[
"Xu",
"Pengxiang",
""
],
[
"Tian",
"Yonghong",
""
]
] |
2507.23702
|
Thien Duc Hua
|
Duc Thien Hua, Mohammadali Mohammadi, Hien Quoc Ngo, Michail Matthaiou
|
Cell-Free Massive MIMO SWIPT with Beyond Diagonal Reconfigurable Intelligent Surfaces
|
This work is the journal version of the article arXiv:2402.00646v1 and identical to arXiv:2402.00646v2
| null |
10.1109/TCOMM.2025.3597660
|
TCOM-TLS-25-0030
|
cs.IT eess.SP math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
We investigate the integration of beyond diagonal reconfigurable intelligent surfaces (BDRISs) into cell free massive multiple input multiple output (CFmMIMO) systems to enhance simultaneous wireless information and power transfer (SWIPT). To simultaneously support two groups of users energy receivers (ERs) and information receivers (IRs) without sacrificing time frequency resources, a subset of access points (APs) is dedicated to serving ERs with the aid of a BDRIS, while the remaining APs focus on supporting IRs. A protective partial zero forcing precoding technique is implemented at the APs to manage the non coherent interference between the ERs and IRs. Subsequently, closed form expressions for the spectral efficiency of the IRs and the average sum of harvested energy at the ERs are leveraged to formulate a comprehensive optimization problem. This problem jointly optimizes the AP selection, AP power control, and scattering matrix design at the BDRIS, all based on long term statistical channel state information. This challenging problem is then effectively transformed into more tractable forms. To solve these sub problems, efficient algorithms are proposed, including a heuristic search for the scattering matrix design, as well as successive convex approximation and deep reinforcement learning methods for the joint AP mode selection and power control design. Numerical results show that a BDRIS with a group or fully connected architecture achieves significant energy harvesting gains over the conventional diagonal RIS, especially delivering up to a seven fold increase in the average sum of harvested energy when a heuristic based scattering matrix design is employed.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 16:24:28 GMT"
},
{
"version": "v2",
"created": "Sun, 03 Aug 2025 00:38:43 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 10:51:59 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hua",
"Duc Thien",
""
],
[
"Mohammadi",
"Mohammadali",
""
],
[
"Ngo",
"Hien Quoc",
""
],
[
"Matthaiou",
"Michail",
""
]
] |
2507.23755
|
Rongzhen Zhao
|
Rongzhen Zhao, Yi Zhao, Juho Kannala, Joni Pajarinen
|
Slot Attention with Re-Initialization and Self-Distillation
|
Accepted by ACM MM 2025
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Unlike popular solutions based on dense feature maps, Object-Centric Learning (OCL) represents visual scenes as sub-symbolic object-level feature vectors, termed slots, which are highly versatile for tasks involving visual modalities. OCL typically aggregates object superpixels into slots by iteratively applying competitive cross attention, known as Slot Attention, with the slots as the query. However, once initialized, these slots are reused naively, causing redundant slots to compete with informative ones for representing objects. This often results in objects being erroneously segmented into parts. Additionally, mainstream methods derive supervision signals solely from decoding slots into the input's reconstruction, overlooking potential supervision based on internal information. To address these issues, we propose Slot Attention with re-Initialization and self-Distillation (DIAS): $\emph{i)}$ We reduce redundancy in the aggregated slots and re-initialize extra aggregation to update the remaining slots; $\emph{ii)}$ We drive the bad attention map at the first aggregation iteration to approximate the good at the last iteration to enable self-distillation. Experiments demonstrate that DIAS achieves state-of-the-art on OCL tasks like object discovery and recognition, while also improving advanced visual prediction and reasoning. Our source code and model checkpoints are available on https://github.com/Genera1Z/DIAS.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 17:41:18 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 16:24:31 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhao",
"Rongzhen",
""
],
[
"Zhao",
"Yi",
""
],
[
"Kannala",
"Juho",
""
],
[
"Pajarinen",
"Joni",
""
]
] |
2508.00875
|
Kevin Wei
|
Patricia Paskov, Michael J. Byun, Kevin Wei, Toby Webster
|
Preliminary suggestions for rigorous GPAI model evaluations
|
Santa Monica, CA: RAND Corporation, 2025. Published as a RAND expert commentary at: https://www.rand.org/pubs/perspectives/PEA3971-1.html
| null |
10.7249/PEA3971-1
|
PE-A3971-1
|
cs.CY cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This document presents a preliminary compilation of general-purpose AI (GPAI) evaluation practices that may promote internal validity, external validity and reproducibility. It includes suggestions for human uplift studies and benchmark evaluations, as well as cross-cutting suggestions that may apply to many different evaluation types. Suggestions are organised across four stages in the evaluation life cycle: design, implementation, execution and documentation. Drawing from established practices in machine learning, statistics, psychology, economics, biology and other fields recognised to have important lessons for AI evaluation, these suggestions seek to contribute to the conversation on the nascent and evolving field of the science of GPAI evaluations. The intended audience of this document includes providers of GPAI models presenting systemic risk (GPAISR), for whom the EU AI Act lays out specific evaluation requirements; third-party evaluators; policymakers assessing the rigour of evaluations; and academic researchers developing or conducting GPAI evaluations.
|
[
{
"version": "v1",
"created": "Tue, 22 Jul 2025 03:27:42 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Paskov",
"Patricia",
""
],
[
"Byun",
"Michael J.",
""
],
[
"Wei",
"Kevin",
""
],
[
"Webster",
"Toby",
""
]
] |
2508.01139
|
Guang Li
|
Huyu Wu, Duo Su, Junjie Hou, Guang Li
|
Dataset Condensation with Color Compensation
| null | null | null | null |
cs.CV cs.AI cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dataset condensation always faces a constitutive trade-off: balancing performance and fidelity under extreme compression. Existing methods struggle with two bottlenecks: image-level selection methods (Coreset Selection, Dataset Quantization) suffer from inefficiency condensation, while pixel-level optimization (Dataset Distillation) introduces semantic distortion due to over-parameterization. With empirical observations, we find that a critical problem in dataset condensation is the oversight of color's dual role as an information carrier and a basic semantic representation unit. We argue that improving the colorfulness of condensed images is beneficial for representation learning. Motivated by this, we propose DC3: a Dataset Condensation framework with Color Compensation. After a calibrated selection strategy, DC3 utilizes the latent diffusion model to enhance the color diversity of an image rather than creating a brand-new one. Extensive experiments demonstrate the superior performance and generalization of DC3 that outperforms SOTA methods across multiple benchmarks. To the best of our knowledge, besides focusing on downstream tasks, DC3 is the first research to fine-tune pre-trained diffusion models with condensed datasets. The FID results prove that training networks with our high-quality datasets is feasible without model collapse or other degradation issues. Code and generated data are available at https://github.com/528why/Dataset-Condensation-with-Color-Compensation.
|
[
{
"version": "v1",
"created": "Sat, 02 Aug 2025 01:44:23 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 02:05:53 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wu",
"Huyu",
""
],
[
"Su",
"Duo",
""
],
[
"Hou",
"Junjie",
""
],
[
"Li",
"Guang",
""
]
] |
2508.01339
|
Chuanqi Liang
|
Chuanqi Liang, Jie Fu, Miao Yu, Lei Luo
|
SBP-YOLO:A Lightweight Real-Time Model for Detecting Speed Bumps and Potholes
|
14pages,10figures
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Reliable and real-time detection of road speed bumps and potholes is crucial for anticipatory perception in advanced suspension systems, enabling timely and adaptive damping control. Achieving high accuracy and efficiency on embedded platforms remains challenging due to limited computational resources and the small scale of distant targets. This paper presents SBP-YOLO, a lightweight and high-speed detection framework tailored for bump and pothole recognition. Based on YOLOv11n, the model integrates GhostConv and VoVGSCSPC modules into the backbone and neck to reduce computation while enhancing multi-scale semantic features. To improve small-object detection, a P2-level branch is introduced with a lightweight and efficient detection head LEDH mitigating the added computational overhead without compromising accuracy. A hybrid training strategy combining NWD loss, backbone-level knowledge distillation, and Albumentations-driven augmentation further enhances localization precision and robustness. Experiments show that SBP-YOLO achieves 87.0 percent mAP, outperforming the YOLOv11n baseline by 5.8 percent. After TensorRT FP16 quantization, it runs at 139.5 FPS on Jetson AGX Xavier, delivering a 12.4 percent speedup over the P2-enhanced YOLOv11. These results validate the effectiveness of the proposed method for fast and low-latency road condition perception in embedded suspension control systems.
|
[
{
"version": "v1",
"created": "Sat, 02 Aug 2025 12:15:08 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:32:27 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Liang",
"Chuanqi",
""
],
[
"Fu",
"Jie",
""
],
[
"Yu",
"Miao",
""
],
[
"Luo",
"Lei",
""
]
] |
2508.01702
|
B.Sundar Rajan
|
Hareesh K., Rashid Ummer N.T., and B. Sundar Rajan
|
Explicit Function-Correcting Code Constructions for Lee Metric Channels
|
The Plotkin-like bound for irregular Lee-distance codes has been revised. Accordingly, the lower bounds on the redundancy of function-correcting Lee codes for specific functions, which rely on this Plotkin-like bound, have also been modified. 16 pages, 4 tables and 1 figure
| null | null | null |
cs.IT math.IT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Function-Correcting Codes (FCCs) are a novel class of codes designed to protect function evaluations of messages against errors while minimizing redundancy. A theoretical framework for systematic FCCs to channels matched to the Lee metric has been studied recently, which introduced function-correcting Lee codes (FCLCs) and also derived upper and lower bounds on their optimal redundancy. In this paper, we first propose a Plotkin-like bound for irregular Lee-distance codes. We then construct explicit FCLCs for specific classes of functions, including the Lee weight, Lee weight distribution, modular sum, and locally bounded function. For these functions, lower bounds on redundancy are obtained, and our constructions are shown to be optimal in certain cases. Finally, a comparative analysis with classical Lee error-correcting codes and codes correcting errors in function values, demonstrates that FCLCs can significantly reduce redundancy while preserving function correctness.
|
[
{
"version": "v1",
"created": "Sun, 03 Aug 2025 10:06:42 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:31:59 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"K.",
"Hareesh",
""
],
[
"T.",
"Rashid Ummer N.",
""
],
[
"Rajan",
"B. Sundar",
""
]
] |
2508.01759
|
Shao-Jiang Wang
|
Jia-Qi Wang, Rong-Gen Cai, Zong-Kuan Guo, Shao-Jiang Wang
|
Resolving the Planck-DESI tension by non-minimally coupled quintessence
|
v1, 5 pages + 6 pages Supplemental Materials, 8 figures, 2 tables; v2, a subscript typo in Eq.(8) is corrected, references are added
| null | null | null |
astro-ph.CO hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Planck measurement of the cosmic microwave background (CMB) has established the $\Lambda$-cold-dark-matter ($\Lambda$CDM) model as the concordant model along with other observations. However, recent measurements of baryon acoustic oscillations (BAO) from the Dark Energy Spectroscopic Instrument (DESI) collaboration have renewed the matter fraction $\Omega_\mathrm{m}$ tension between DESI-$\Lambda$CDM and Planck-$\Lambda$CDM. Directly reconciling this CMB-BAO tension with a dynamical DE in Chevallier-Polarski-Linder (CPL) parametrization seems to imply a crossing of the equation-of-state through $w=-1$ at low redshifts. In this letter, we will illustrate with a string-theory-motivated model that, when the DM non-minimally couples to gravity via a quintessence field, a misidentification with the $w_0w_a$CDM model would exactly fake such a crossing behavior, while the coupled quintessence never crosses $w=-1$ but behaves as a standard CDM in the early Universe and approaches a cosmological constant in the late Universe. Such a non-minimal coupling is preferred over $3\sigma$ confidence level. The worsened $\Omega_\mathrm{m}$ tension and $S_8$ tension in the $w_0w_a$CDM model are also resolved in our model.
|
[
{
"version": "v1",
"created": "Sun, 03 Aug 2025 14:00:54 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 16:08:58 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Jia-Qi",
""
],
[
"Cai",
"Rong-Gen",
""
],
[
"Guo",
"Zong-Kuan",
""
],
[
"Wang",
"Shao-Jiang",
""
]
] |
2508.02098
|
Shamik Banerjee
|
Shamik Banerjee, Mousumi Maitra, Raju Mandal, Milan Patra
|
Holographic symmetry algebra for the MHV sector revisited
|
18 Pages, Latex, Author added
| null | null | null |
hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
We revisit the holographic symmetry algebra in the MHV sector. We find an infinite dimensional Abelian symmetry algebra whose generators are the conformally soft negative helicity gravitons and gluons. So the complete symmetry algebra in the MHV graviton sector is a semideirect product of the $w_{1+\infty}$ algebra and the infinite dimensional Abelian algebra. Similarly in the MHV gluon sector the symmetry algebra is a semidirect product of the $S$ algebra and the infinite dimensional Abelian algebra. The extended symmetry algebra has some use. For example, it is known for sometime that an $n$ point MHV amplitude satisfies $(n-2)$ Knizhnik-Zamolodchikov (KZ) type equations. So two equations are missing. We show that the extended symmetry algebra has additional null states whose decoupling give rise to the two missing equations.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 06:12:23 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 14:05:46 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 03:58:36 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Banerjee",
"Shamik",
""
],
[
"Maitra",
"Mousumi",
""
],
[
"Mandal",
"Raju",
""
],
[
"Patra",
"Milan",
""
]
] |
2508.02262
|
Makoto Ishihara
|
Makoto Ishihara, Anthony Brendan, Wojciech Roga, Ulrik L. Andersen, Masahiro Takeoka
|
Long-distance device-independent quantum key distribution with standard optics tools
|
11 pages, 7 figures
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Device-independent quantum key distribution (DI-QKD) enables information-theoretically secure key exchange between remote parties without any assumptions on the internal workings of the devices used for its implementation. However, its practical deployment remains severely constrained by the need for loophole-free Bell inequality violations, which are highly susceptible to losses and detection efficiencies. In this paper, we propose two long-distance DI-QKD protocols based on a heralding scheme using single-photon interference. Our protocols consist of only standard quantum optics tools such as two-mode squeezed states, displacement operations and on-off detectors, making them experimentally accessible. To further enhance robustness against realistic imperfections, we integrate a classical noisy preprocessing technique during post-processing. We calculate key rates of the protocols by numerical optimization and show the supremacy of this implementation over existing protocols in terms of communication distances.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 10:10:59 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 06:25:29 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ishihara",
"Makoto",
""
],
[
"Brendan",
"Anthony",
""
],
[
"Roga",
"Wojciech",
""
],
[
"Andersen",
"Ulrik L.",
""
],
[
"Takeoka",
"Masahiro",
""
]
] |
2508.02350
|
Abhishek Dhar Dr.
|
Abhishek Dhar, Sarthak Mishra, Spandan Roy and Daniel Axehill
|
Adaptive Lattice-based Motion Planning
| null | null | null | null |
cs.RO cs.SY eess.SY
|
http://creativecommons.org/licenses/by/4.0/
|
This paper proposes an adaptive lattice-based motion planning solution to address the problem of generating feasible trajectories for systems, represented by a linearly parameterizable non-linear model operating within a cluttered environment. The system model is considered to have uncertain model parameters. The key idea here is to utilize input/output data online to update the model set containing the uncertain system parameter, as well as a dynamic estimated parameter of the model, so that the associated model estimation error reduces over time. This in turn improves the quality of the motion primitives generated by the lattice-based motion planner using a nominal estimated model selected on the basis of suitable criteria. The motion primitives are also equipped with tubes to account for the model mismatch between the nominal estimated model and the true system model, to guarantee collision-free overall motion. The tubes are of uniform size, which is directly proportional to the size of the model set containing the uncertain system parameter. The adaptive learning module guarantees a reduction in the diameter of the model set as well as in the parameter estimation error between the dynamic estimated parameter and the true system parameter. This directly implies a reduction in the size of the implemented tubes and guarantees that the utilized motion primitives go arbitrarily close to the resolution-optimal motion primitives associated with the true model of the system, thus significantly improving the overall motion planning performance over time. The efficiency of the motion planner is demonstrated by a suitable simulation example that considers a drone model represented by Euler-Lagrange dynamics containing uncertain parameters and operating within a cluttered environment.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 12:35:07 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:21:28 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dhar",
"Abhishek",
""
],
[
"Mishra",
"Sarthak",
""
],
[
"Roy",
"Spandan",
""
],
[
"Axehill",
"Daniel",
""
]
] |
2508.02619
|
Narendra Nimmala
|
Dilip Kumar Ghosh, Anish Ghoshal, Koustav Mukherjee, Nimmala Narendra, Nobuchika Okada
|
Impact of Non-Thermal Leptogenesis with Early Matter Domination on Gravitational Waves from First-order Phase Transition
|
32 pages, 5 figures, Added citations
| null | null | null |
hep-ph astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the impact of non-thermal leptogenesis on the spectrum of gravitational waves (GWs) produced by a strong first-order phase transition in the early Universe. We consider a scenario in which a heavy scalar field, $\phi$, dominates the energy density of the early Universe and decays into heavy right-handed neutrinos (RHNs). The subsequent decay of RHNs generates a lepton asymmetry, which is partially converted into the observed baryon asymmetry via the sphaleron process. The $\phi$-dominated era and the entropy injection from the decays of $\phi$ and RHNs leave characteristic imprints on the GW spectrum, such as damping and modified frequency dependence, that distinguish it from the standard cosmological evolution. We identify the parameter space in which non-thermal leptogenesis is successful, leading to distinctive GW spectral features. We show that these GW signals can fall within the sensitivity ranges of future detectors such as ET, DECIGO and BBO. If observed, they would provide valuable insights into the thermal history and dynamics of the early Universe.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 17:06:37 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 09:38:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ghosh",
"Dilip Kumar",
""
],
[
"Ghoshal",
"Anish",
""
],
[
"Mukherjee",
"Koustav",
""
],
[
"Narendra",
"Nimmala",
""
],
[
"Okada",
"Nobuchika",
""
]
] |
2508.02909
|
Aakash Gore
|
Aakash Gore and Prasanna Chaporkar
|
Clus-UCB: A Near-Optimal Algorithm for Clustered Bandits
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
We study a stochastic multi-armed bandit setting where arms are partitioned into known clusters, such that the mean rewards of arms within a cluster differ by at most a known threshold. While the clustering structure is known a priori, the arm means are unknown. We derive an asymptotic lower bound on the regret that improves upon the classical bound of Lai & Robbins (1985). We then propose Clus-UCB, an efficient algorithm that closely matches this lower bound asymptotically. Clus-UCB is designed to exploit the clustering structure and introduces a new index to evaluate an arm, which depends on other arms within the cluster. In this way, arms share information among each other. We present simulation results of our algorithm and compare its performance against KL-UCB and other wellknown algorithms for bandits with dependent arms. Finally, we address some limitations of this work and conclude by mentioning some possible future research.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 21:21:59 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 15:15:15 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Gore",
"Aakash",
""
],
[
"Chaporkar",
"Prasanna",
""
]
] |
2508.02946
|
Francisco Rouxinol
|
Italo L. Soares Andrade, Kleber Pirota, Amir O. Caldeira and Francisco Rouxinol
|
Observation of Purcell Effect in Electrically Coupled Cavity-Magnet System
| null | null | null | null |
quant-ph cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We report the observation of the Purcell effect in a cavity-metallic magnet hybrid system using electric-field-mediated coupling. In this configuration, microwave-induced axial currents in the microwire induce circular magnetic fields that drive the ferromagnetic resonance (FMR) of the magnetized microwire. Field-dependent transmission and reflection spectroscopies reveal a clear cavity perturbation consistent with the Purcell regime, in which the magnetic loss rate exceeds the light-matter coupling strength. Despite the small magnetic volume ($\sim 10^{-13}\,\text{m}^3$), measurements performed at both room temperature and $T = 7$ mK show coupling rates as high as $g/2pi = 56$ MHz, one order of magnitude stronger than the one expected from conventional coupling at the magnetic antinode. Time-domain ringdown measurements directly show the magnetic-field-dependent modification of the cavity photon lifetime, in agreement with theoretical predictions. These results establish a versatile approach for coupling microwave fields to metallic magnets via geometric and electric-field-mediated interactions, opening new opportunities for hybrid cavity-magnet systems.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 23:04:17 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 17:10:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Andrade",
"Italo L. Soares",
""
],
[
"Pirota",
"Kleber",
""
],
[
"Caldeira",
"Amir O.",
""
],
[
"Rouxinol",
"Francisco",
""
]
] |
2508.03411
|
Radu Tudor Ionescu
|
Diana-Nicoleta Grigore, Neelu Madan, Andreas Mogelmose, Thomas B. Moeslund, Radu Tudor Ionescu
|
SlotMatch: Distilling Temporally Consistent Object-Centric Representations for Unsupervised Video Segmentation
| null | null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Unsupervised video segmentation is a challenging computer vision task, especially due to the lack of supervisory signals coupled with the complexity of visual scenes. To overcome this challenge, state-of-the-art models based on slot attention often have to rely on large and computationally expensive neural architectures. To this end, we propose a simple knowledge distillation framework that effectively transfers object-centric representations to a lightweight student. The proposed framework, called SlotMatch, aligns corresponding teacher and student slots via the cosine similarity, requiring no additional distillation objectives or auxiliary supervision. The simplicity of SlotMatch is confirmed via theoretical and empirical evidence, both indicating that integrating additional losses is redundant. We conduct experiments on two datasets to compare the state-of-the-art teacher model, SlotContrast, with our distilled student. The results show that our student based on SlotMatch matches and even outperforms its teacher, while using 3.6x less parameters and running 1.9x faster. Moreover, our student surpasses previous unsupervised video segmentation models.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 12:58:09 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:24:07 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Grigore",
"Diana-Nicoleta",
""
],
[
"Madan",
"Neelu",
""
],
[
"Mogelmose",
"Andreas",
""
],
[
"Moeslund",
"Thomas B.",
""
],
[
"Ionescu",
"Radu Tudor",
""
]
] |
2508.04107
|
JingChao Wang
|
Jingchao Wang, Zhijian Wu, Dingjiang Huang, Yefeng Zheng, Hong Wang
|
Unlocking the Potential of MLLMs in Referring Expression Segmentation via a Light-weight Mask Decoder
|
9 pages, 4 figures
| null | null | null |
cs.CV cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reference Expression Segmentation (RES) aims to segment image regions specified by referring expressions and has become popular with the rise of multimodal large models (MLLMs). While MLLMs excel in semantic understanding, their token-generation paradigm struggles with pixel-level dense prediction. Existing RES methods either couple MLLMs with the parameter-heavy Segment Anything Model (SAM) with 632M network parameters or adopt SAM-free lightweight pipelines that sacrifice accuracy. To address the trade-off between performance and cost, we specifically propose MLLMSeg, a novel framework that fully exploits the inherent visual detail features encoded in the MLLM vision encoder without introducing an extra visual encoder. Besides, we propose a detail-enhanced and semantic-consistent feature fusion module (DSFF) that fully integrates the detail-related visual feature with the semantic-related feature output by the large language model (LLM) of MLLM. Finally, we establish a light-weight mask decoder with only 34M network parameters that optimally leverages detailed spatial features from the visual encoder and semantic features from the LLM to achieve precise mask prediction. Extensive experiments demonstrate that our method generally surpasses both SAM-based and SAM-free competitors, striking a better balance between performance and cost. Code is available at https://github.com/jcwang0602/MLLMSeg.
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 06:06:52 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Aug 2025 05:07:54 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 08:35:04 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Jingchao",
""
],
[
"Wu",
"Zhijian",
""
],
[
"Huang",
"Dingjiang",
""
],
[
"Zheng",
"Yefeng",
""
],
[
"Wang",
"Hong",
""
]
] |
2508.04386
|
Leslie Molag
|
J. Marzo, L. D. Molag and J. Ortega-Cerd\`a
|
Universality for fluctuations of counting statistics of random normal matrices
|
31 pages
| null | null | null |
math.PR math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We consider the fluctuations of the number of eigenvalues of $n\times n$ random normal matrices depending on a potential $Q$ in a given set $A$. These eigenvalues are known to form a determinantal point process, and are known to accumulate on a compact set called the droplet under mild conditions on $Q$. When $A$ is a Borel set strictly inside the droplet, we show that the variance of the number of eigenvalues $N_A^{(n)}$ in $A$ has a limiting behavior given by
\begin{align*} \lim_{n\to\infty} \frac1{\sqrt n}\operatorname{Var } N_A^{(n)} = \frac{1}{2\pi\sqrt\pi}\int_{\partial_* A} \sqrt{\Delta Q(z)} \, d\mathcal H^1(z), \end{align*} where $\partial_* A$ is the measure theoretic boundary of $A$, $d\mathcal H^1(z)$ denotes the one-dimensional Hausdorff measure, and $\Delta = \partial_z \overline{\partial_z}$. We also consider the case where $A$ is a microscopic dilation of the droplet and fully generalize a result by Akemann, Byun and Ebke for arbitrary potentials. In this result $d\mathcal H^1(z)$ is replaced by the harmonic measure at $\infty$ associated with the exterior of the droplet. This second result is proved by strengthening results due to Hedenmalm-Wennman and Ameur-Cronvall on the asymptotic behavior of the associated correlation kernel near the droplet boundary.
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 12:29:32 GMT"
},
{
"version": "v2",
"created": "Thu, 07 Aug 2025 04:14:52 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 16:55:41 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Marzo",
"J.",
""
],
[
"Molag",
"L. D.",
""
],
[
"Ortega-Cerdà",
"J.",
""
]
] |
2508.04880
|
Siddharth Dangwal
|
Siddharth Dangwal, Tina Oberoi, Ajay Sailopal, Dhirpal Shah, Frederic T. Chong
|
Noisy Quantum Simulation Using Tracking, Uncomputation and Sampling
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum computers have grown in size and qubit quality in recent years, enabling the execution of complex quantum circuits. However, for most researchers, access to compute time on quantum hardware is limited. This necessitates the need to build simulators that mimic the execution of quantum circuits on noisy quantum hardware accurately and scalably.
In this work, we propose TUSQ - Tracking, Uncomputation, and Sampling for Noisy Quantum Simulation. To represent the stochastic noisy channels accurately, we average the output of multiple quantum circuits with fixed noisy gates sampled from the channels. However, this leads to a substantial increase in circuit overhead, which slows down the simulation. To eliminate this overhead, TUSQ uses two modules: the Error Characterization Module (ECM), and the Tree-based Execution Module (TEM).
The ECM tracks the number of unique circuit executions needed to accurately represent the noise. That is, if initially we needed $n_{1}$ circuit executions, ECM reduces that number to $n_{2}$ by eliminating redundancies so that $n_{2} < n_{1}$. This is followed by the TEM, which reuses computation across these $n_{2}$ circuits. This computational reuse is facilitated by representing all $n_{2}$ circuits as a tree. We sample the significant leaf nodes of this tree and prune the remaining ones. We traverse this tree using depth-first search. We use uncomputation to perform rollback-recovery at several stages which reduces simulation time. We evaluate TUSQ for a total of 186 benchmarks and report an average speedup of $52.5\times$ and $12.53\times$ over Qiskit and CUDA-Q, which goes up to $7878.03\times$ and $439.38\times$ respectively. For larger benchmarks (more than than 15 qubits), the average speedup is $55.42\times$ and $23.03\times$ over Qiskit and CUDA-Q respectively
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 21:08:39 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 18:33:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dangwal",
"Siddharth",
""
],
[
"Oberoi",
"Tina",
""
],
[
"Sailopal",
"Ajay",
""
],
[
"Shah",
"Dhirpal",
""
],
[
"Chong",
"Frederic T.",
""
]
] |
2508.05064
|
Todd Charter
|
Mahmoud Chick Zaouali, Todd Charter, Yehor Karpichev, Brandon Haworth, Homayoun Najjaran
|
A Study of the Framework and Real-World Applications of Language Embedding for 3D Scene Understanding
| null | null | null | null |
cs.GR cs.CL cs.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Gaussian Splatting has rapidly emerged as a transformative technique for real-time 3D scene representation, offering a highly efficient and expressive alternative to Neural Radiance Fields (NeRF). Its ability to render complex scenes with high fidelity has enabled progress across domains such as scene reconstruction, robotics, and interactive content creation. More recently, the integration of Large Language Models (LLMs) and language embeddings into Gaussian Splatting pipelines has opened new possibilities for text-conditioned generation, editing, and semantic scene understanding. Despite these advances, a comprehensive overview of this emerging intersection has been lacking. This survey presents a structured review of current research efforts that combine language guidance with 3D Gaussian Splatting, detailing theoretical foundations, integration strategies, and real-world use cases. We highlight key limitations such as computational bottlenecks, generalizability, and the scarcity of semantically annotated 3D Gaussian data and outline open challenges and future directions for advancing language-guided 3D scene understanding using Gaussian Splatting.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 06:33:08 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 00:47:37 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zaouali",
"Mahmoud Chick",
""
],
[
"Charter",
"Todd",
""
],
[
"Karpichev",
"Yehor",
""
],
[
"Haworth",
"Brandon",
""
],
[
"Najjaran",
"Homayoun",
""
]
] |
2508.05068
|
Ruiyu Li
|
Changyuan Qiu, Hangrui Cao, Qihan Ren, Ruiyu Li, Yuqing Qiu
|
Automatic Image Colorization with Convolutional Neural Networks and Generative Adversarial Networks
|
All authors have equal authorship and equal contribution, ranked in alphabetic order. First version of this paper was completed and published in 2021
| null | null | null |
cs.CV cs.AI cs.LG eess.IV
|
http://creativecommons.org/licenses/by/4.0/
|
Image colorization, the task of adding colors to grayscale images, has been the focus of significant research efforts in computer vision in recent years for its various application areas such as color restoration and automatic animation colorization [15, 1]. The colorization problem is challenging as it is highly ill-posed with two out of three image dimensions lost, resulting in large degrees of freedom. However, semantics of the scene as well as the surface texture could provide important cues for colors: the sky is typically blue, the clouds are typically white and the grass is typically green, and there are huge amounts of training data available for learning such priors since any colored image could serve as a training data point [20].
Colorization is initially formulated as a regression task[5], which ignores the multi-modal nature of color prediction. In this project, we explore automatic image colorization via classification and adversarial learning. We will build our models on prior works, apply modifications for our specific scenario and make comparisons.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 06:41:31 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:06:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Qiu",
"Changyuan",
""
],
[
"Cao",
"Hangrui",
""
],
[
"Ren",
"Qihan",
""
],
[
"Li",
"Ruiyu",
""
],
[
"Qiu",
"Yuqing",
""
]
] |
2508.05287
|
Lars Graf
|
Lars Graf, Thomas Ortner, Stanis{\l}aw Wo\'zniak, Angeliki Pantazi
|
FlowState: Sampling Rate Invariant Time Series Forecasting
|
Currently under review
| null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Foundation models (FMs) have transformed natural language processing, but their success has not yet translated to time series forecasting. Existing time series foundation models (TSFMs), often based on transformer variants, struggle with generalization across varying context and target lengths, lack adaptability to different sampling rates, and are computationally inefficient. We introduce FlowState, a novel TSFM architecture that addresses these challenges through two key innovations: a state space model (SSM) based encoder and a functional basis decoder. This design enables continuous-time modeling and dynamic time-scale adjustment, allowing FlowState to inherently generalize across all possible temporal resolutions, and dynamically adjust the forecasting horizons. In contrast to other state-of-the-art TSFMs, which require training data across all possible sampling rates to memorize patterns at each scale, FlowState inherently adapts its internal dynamics to the input scale, enabling smaller models, reduced data requirements, and improved efficiency. We further propose an efficient pretraining strategy that improves robustness and accelerates training. Despite being the smallest model, FlowState outperforms all other models and is state-of-the-art for the GIFT-ZS and the Chronos-ZS benchmarks. Ablation studies confirm the effectiveness of its components, and we demonstrate its unique ability to adapt online to varying input sampling rates.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 11:30:26 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 12:17:08 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Graf",
"Lars",
""
],
[
"Ortner",
"Thomas",
""
],
[
"Woźniak",
"Stanisław",
""
],
[
"Pantazi",
"Angeliki",
""
]
] |
2508.05413
|
Andrei Pavlov I
|
Andrei I. Pavlov, Mikhail N. Kiselev
|
Universal relations between thermoelectrics and noise in mesoscopic transport across a tunnel junction
| null | null | null | null |
cond-mat.mes-hall cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We develop a unified theory of weakly probed differential observables for currents and noise in transport experiments. Our findings uncover a set of universal transport relations between thermoelectric and noise properties of a system probed through a tunnel contact, with the Wiedemann-Franz law being just one example of such universality between charge and heat currents. We apply this theory to various quantum dot systems, including multichannel Kondo, quantum Hall and Sachdev-Ye-Kitaev quantum dots, and demonstrate that each of the microscopic theories is characterized by a set of universal relations connecting conductance and thermoelectrics with noise. Violations of these relations indicate additional energy scales emerging in a system.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 14:06:57 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 12:29:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Pavlov",
"Andrei I.",
""
],
[
"Kiselev",
"Mikhail N.",
""
]
] |
2508.05597
|
Simon Mackenzie Dr
|
Serge Gaspers, Tao Zixu He, Simon Mackenzie
|
NP-Hardness and ETH-Based Inapproximability of Communication Complexity via Relaxed Interlacing
|
v3 contains proof sketches for the ETH gap; full details forthcoming. v2 corrects a misstatement in v1 about the concurrent work (Hirahara-Llango-Loff)
| null | null | null |
cs.CC cs.DS math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove that computing the deterministic communication complexity D(f) of a Boolean function is NP-hard in the standard protocol-tree model, answering, independently and concurrently with Hirahara-Llango-Loff (arXiv:2507.10426), a question first posed by Yao (1979). Our reduction builds and expands on a suite of structural "interlacing" lemmas introduced by Mackenzie and Saffidine (arXiv:2411.19003); these lemmas can be reused as black boxes in future lower-bound constructions.
The instances produced by our reduction admit optimal protocols for self-similar constructions with strong structural properties, giving a flexible framework for the design of reductions showing NP-hardness of deciding the communication complexity of a Boolean matrix. This complements the work by Hirahara, Ilango, and Loff, which establishes NP-hardness in the same model via a different route; our analysis additionally yields reusable structural guarantees and underpins further consequences concerning inapproximability.
Because the gadgets in our construction are self-similar, they can be recursively embedded. We sketch how this yields, under the Exponential-Time Hypothesis, an additive inapproximability gap that grows without bound. Furthermore we outline a route toward NP-hardness of approximating D(f) within a fixed constant additive error. Full details of the ETH-based inapproximability results will appear in a future version.
Beyond settling the complexity of deterministic communication complexity itself, the modular framework we develop opens the door to a wider class of reductions and, we believe, will prove useful in tackling other long-standing questions in communication complexity.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 17:39:40 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 12:12:12 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 13:29:52 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Gaspers",
"Serge",
""
],
[
"He",
"Tao Zixu",
""
],
[
"Mackenzie",
"Simon",
""
]
] |
2508.05668
|
Yunjia Xi
|
Yunjia Xi, Jianghao Lin, Yongzhao Xiao, Zheli Zhou, Rong Shan, Te Gao, Jiachen Zhu, Weiwen Liu, Yong Yu, Weinan Zhang
|
A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges
| null | null | null | null |
cs.IR cs.AI cs.CL
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The advent of Large Language Models (LLMs) has significantly revolutionized web search. The emergence of LLM-based Search Agents marks a pivotal shift towards deeper, dynamic, autonomous information seeking. These agents can comprehend user intentions and environmental context and execute multi-turn retrieval with dynamic planning, extending search capabilities far beyond the web. Leading examples like OpenAI's Deep Research highlight their potential for deep information mining and real-world applications. This survey provides the first systematic analysis of search agents. We comprehensively analyze and categorize existing works from the perspectives of architecture, optimization, application, and evaluation, ultimately identifying critical open challenges and outlining promising future research directions in this rapidly evolving field. Our repository is available on https://github.com/YunjiaXi/Awesome-Search-Agent-Papers.
|
[
{
"version": "v1",
"created": "Sun, 03 Aug 2025 08:02:51 GMT"
},
{
"version": "v2",
"created": "Sun, 17 Aug 2025 13:52:26 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 05:15:19 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Xi",
"Yunjia",
""
],
[
"Lin",
"Jianghao",
""
],
[
"Xiao",
"Yongzhao",
""
],
[
"Zhou",
"Zheli",
""
],
[
"Shan",
"Rong",
""
],
[
"Gao",
"Te",
""
],
[
"Zhu",
"Jiachen",
""
],
[
"Liu",
"Weiwen",
""
],
[
"Yu",
"Yong",
""
],
[
"Zhang",
"Weinan",
""
]
] |
2508.05796
|
Valentina Bisogni
|
Yanhong Gu, Joseph Barker, Jiemin Li, Takashi Kikkawa, Fernando Camino, Kim Kisslinger, John Sinsheimer, Lukas Lienhard, Jackson J. Bauer, Caroline A. Ross, Dmitri N. Basov, Eiji Saitoh, Jonathan Pelliciari, Gerrit E. W. Bauer and Valentina Bisogni
|
Observing Differential Spin Currents by Resonant Inelastic X-ray Scattering
| null | null | null | null |
cond-mat.mes-hall cond-mat.mtrl-sci cond-mat.str-el
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Controlling spin currents, i.e., the flow of spin angular momentum, in small magnetic devices is the principal objective of spin electronics, a main contender for future energy efficient information technologies. Surprisingly, a pure spin current has never been measured directly since the associated electric stray fields and/or shifts in the non-equilibrium spin-dependent distribution functions are too small for conventional experimental detection methods optimized for charge transport. Here we report that resonant inelastic x-ray scattering (RIXS) can bridge this gap by measuring the spin current carried by magnons -- the quanta of the spin wave excitations of the magnetic order -- in the presence of temperature gradients across a magnetic insulator. This is possible due to the sensitivity of the momentum- and energy-resolved RIXS intensity to minute changes in the magnon distribution under non-equilibrium conditions. We use the Boltzmann equation in the relaxation time approximation to extract transport parameters, such as the magnon lifetime at finite momentum, essential for the realization of magnon spintronics.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 19:09:53 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 20:14:00 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Gu",
"Yanhong",
""
],
[
"Barker",
"Joseph",
""
],
[
"Li",
"Jiemin",
""
],
[
"Kikkawa",
"Takashi",
""
],
[
"Camino",
"Fernando",
""
],
[
"Kisslinger",
"Kim",
""
],
[
"Sinsheimer",
"John",
""
],
[
"Lienhard",
"Lukas",
""
],
[
"Bauer",
"Jackson J.",
""
],
[
"Ross",
"Caroline A.",
""
],
[
"Basov",
"Dmitri N.",
""
],
[
"Saitoh",
"Eiji",
""
],
[
"Pelliciari",
"Jonathan",
""
],
[
"Bauer",
"Gerrit E. W.",
""
],
[
"Bisogni",
"Valentina",
""
]
] |
2508.05854
|
Javier Pena
|
Javier Pena, Vikesh Siddhu, and Sridhar Tayur
|
Tailored First-order and Interior-point methods and a new semidefinite programming hierarchy for entanglement detection
|
55 pages, 5 figures
| null | null | null |
quant-ph math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum entanglement lies at the heart of quantum information science, yet its reliable detection in high-dimensional or noisy systems remains a fundamental computational challenge. Semidefinite programming (SDP) hierarchies, such as the Doherty-Parrilo-Spedalieri (DPS) and Extension (EXT) hierarchies, offer complete methods for entanglement detection, but their practical use is limited by exponential growth in problem size. In this paper, we introduce a new SDP hierarchy, PST, that is sandwiched between EXT and DPS--offering a tighter approximation to the set of separable states than EXT, while incurring lower computational overhead than DPS.
We develop compact, polynomially-scalable descriptions of EXT and PST using partition mappings and operators. These descriptions in turn yield formulations that satisfy desirable properties such as the Slater condition and are well-suited to both first-order methods (FOMs) and interior-point methods (IPMs). We design a suite of entanglement detection algorithms: three FOMs (Frank-Wolfe, projected gradient, and fast projected gradient) based on a least-squares formulation, and a custom primal-dual IPM based on a conic programming formulation. These methods are numerically stable and capable of producing entanglement witnesses or proximity measures, even in cases where states lie near the boundary of separability.
Numerical experiments on benchmark quantum states demonstrate that our algorithms improve the ability to solve deeper levels of the SDP hierarchy. In particular, the PST hierarchy combined with FOMs enables scalable and effective entanglement detection in relatively easy instances, while our IPM approach offers robustness and early witness recovery for the more difficult ones. Our results highlight the benefits of tailoring algorithmic formulations to hierarchy structure to advance entanglement detection at scale.
|
[
{
"version": "v1",
"created": "Thu, 07 Aug 2025 21:07:36 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:09:27 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Pena",
"Javier",
""
],
[
"Siddhu",
"Vikesh",
""
],
[
"Tayur",
"Sridhar",
""
]
] |
2508.06189
|
Cheng Liu
|
Cheng Liu, Daou Zhang, Tingxu Liu, Yuhan Wang, Jinyang Chen, Yuexuan Li, Xinying Xiao, Chenbo Xin, Ziru Wang, Weichao Wu
|
MA-CBP: A Criminal Behavior Prediction Framework Based on Multi-Agent Asynchronous Collaboration
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
With the acceleration of urbanization, criminal behavior in public scenes poses an increasingly serious threat to social security. Traditional anomaly detection methods based on feature recognition struggle to capture high-level behavioral semantics from historical information, while generative approaches based on Large Language Models (LLMs) often fail to meet real-time requirements. To address these challenges, we propose MA-CBP, a criminal behavior prediction framework based on multi-agent asynchronous collaboration. This framework transforms real-time video streams into frame-level semantic descriptions, constructs causally consistent historical summaries, and fuses adjacent image frames to perform joint reasoning over long- and short-term contexts. The resulting behavioral decisions include key elements such as event subjects, locations, and causes, enabling early warning of potential criminal activity. In addition, we construct a high-quality criminal behavior dataset that provides multi-scale language supervision, including frame-level, summary-level, and event-level semantic annotations. Experimental results demonstrate that our method achieves superior performance on multiple datasets and offers a promising solution for risk warning in urban public safety scenarios.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 10:12:00 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:14:11 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Liu",
"Cheng",
""
],
[
"Zhang",
"Daou",
""
],
[
"Liu",
"Tingxu",
""
],
[
"Wang",
"Yuhan",
""
],
[
"Chen",
"Jinyang",
""
],
[
"Li",
"Yuexuan",
""
],
[
"Xiao",
"Xinying",
""
],
[
"Xin",
"Chenbo",
""
],
[
"Wang",
"Ziru",
""
],
[
"Wu",
"Weichao",
""
]
] |
2508.06395
|
Ying Hu
|
Steven Boyer, Cameron McA. Gordon, Ying Hu
|
Slope detection, taut foliations, and the relative L-space conjecture
|
There is an error in the proof of Theorem 5.7 in the previous version
| null | null | null |
math.GT
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The $L$-space conjecture asserts the equivalence, for prime $3$-manifolds, of three properties: not being an $L$-space ($NLS$), having a left-orderable fundamental group ($LO$), and admitting a co-orientable taut foliation ($CTF$). In this paper we introduce a relative version of the $L$-space conjecture for knot manifolds $M$, stated in terms of sets of slopes on $\partial M$ characterised (i.e. detected) by Heegaard Floer homology, left-orders, and foliations, respectively. We give a unified characterisation of slope detection, and conjecture that the relative $L$-space conjecture is equivalent to the $L$-space conjecture for toroidal manifolds. We confirm this equivalence for the properties $CTF$ and $NLS$.
Much of our technical work lies in proving that the set of $CTF$-detected slopes on $\partial M$ is a finite union of possibly degenerate closed intervals with rational endpoints; in particular, it is closed in the space of slopes. This involves generalizing results of Tao Li on laminar branched surfaces to the setting of manifolds with boundary.
Within the slopes detected by co-orientable taut foliations, we identify a special subset, which we call exceptional $CTF$-detected slopes. This set includes $CTF$-detected slopes whose associated Dehn fillings don't admit co-orientable taut foliations. We believe this exceptional set is important to understand. In this article, we show that the set of exceptional slopes is finite. However, many questions remain open.
Finally, in the last section of the article we provide a structured synthesis of previous work in the area.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 15:27:46 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 21:18:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Boyer",
"Steven",
""
],
[
"Gordon",
"Cameron McA.",
""
],
[
"Hu",
"Ying",
""
]
] |
2508.06627
|
Mosbah Aouad
|
Mosbah Aouad, Anirudh Choudhary, Awais Farooq, Steven Nevers, Lusine Demirkhanyan, Bhrandon Harris, Suguna Pappu, Christopher Gondi, Ravishankar Iyer
|
Early Detection of Pancreatic Cancer Using Multimodal Learning on Electronic Health Records
| null |
Proceedings of Machine Learning for Healthcare (2025)
| null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Pancreatic ductal adenocarcinoma (PDAC) is one of the deadliest cancers, and early detection remains a major clinical challenge due to the absence of specific symptoms and reliable biomarkers. In this work, we propose a new multimodal approach that integrates longitudinal diagnosis code histories and routinely collected laboratory measurements from electronic health records to detect PDAC up to one year prior to clinical diagnosis. Our method combines neural controlled differential equations to model irregular lab time series, pretrained language models and recurrent networks to learn diagnosis code trajectory representations, and cross-attention mechanisms to capture interactions between the two modalities. We develop and evaluate our approach on a real-world dataset of nearly 4,700 patients and achieve significant improvements in AUC ranging from 6.5% to 15.5% over state-of-the-art methods. Furthermore, our model identifies diagnosis codes and laboratory panels associated with elevated PDAC risk, including both established and new biomarkers. Our code is available at https://github.com/MosbahAouad/EarlyPDAC-MML.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 18:18:15 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 01:02:27 GMT"
},
{
"version": "v3",
"created": "Mon, 18 Aug 2025 23:48:23 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Aouad",
"Mosbah",
""
],
[
"Choudhary",
"Anirudh",
""
],
[
"Farooq",
"Awais",
""
],
[
"Nevers",
"Steven",
""
],
[
"Demirkhanyan",
"Lusine",
""
],
[
"Harris",
"Bhrandon",
""
],
[
"Pappu",
"Suguna",
""
],
[
"Gondi",
"Christopher",
""
],
[
"Iyer",
"Ravishankar",
""
]
] |
2508.06634
|
Hong Zhao
|
Hong Zhao, Jin Wei-Kocsis, Adel Heidari Akhijahani, Karen L Butler-Purry
|
Dual-Head Physics-Informed Graph Decision Transformer for Distribution System Restoration
| null | null | null | null |
eess.SY cs.SY
|
http://creativecommons.org/licenses/by/4.0/
|
Driven by recent advances in sensing and computing, deep reinforcement learning (DRL) technologies have shown great potential for addressing distribution system restoration (DSR) under uncertainty. However, their data-intensive nature and reliance on the Markov Decision Process (MDP) assumption limit their ability to handle scenarios that require long-term temporal dependencies or few-shot and zero-shot decision making. Emerging Decision Transformers (DTs), which leverage causal transformers for sequence modeling in DRL tasks, offer a promising alternative. However, their reliance on return-to-go (RTG) cloning and limited generalization capacity restricts their effectiveness in dynamic power system environments. To address these challenges, we introduce an innovative Dual-Head Physics-informed Graph Decision Transformer (DH-PGDT) that integrates physical modeling, structural reasoning, and subgoal-based guidance to enable scalable and robust DSR even in zero-shot or few-shot scenarios. DH-PGDT features a dual-head physics-informed causal transformer architecture comprising Guidance Head, which generates subgoal representations, and Action Head, which uses these subgoals to generate actions independently of RTG. It also incorporates an operational constraint-aware graph reasoning module that encodes power system topology and operational constraints to generate a confidence-weighted action vector for refining DT trajectories. This design effectively improves generalization and enables robust adaptation to unseen scenarios. While this work focuses on DSR, the underlying computing model of the proposed PGDT is broadly applicable to sequential decision making across various power system operations and other complex engineering domains.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 18:32:45 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:00:31 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhao",
"Hong",
""
],
[
"Wei-Kocsis",
"Jin",
""
],
[
"Akhijahani",
"Adel Heidari",
""
],
[
"Butler-Purry",
"Karen L",
""
]
] |
2508.06649
|
Minjia Mao
|
Daniel Wang, Eli Brignac, Minjia Mao, Xiao Fang
|
Measuring Stereotype and Deviation Biases in Large Language Models
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large language models (LLMs) are widely applied across diverse domains, raising concerns about their limitations and potential risks. In this study, we investigate two types of bias that LLMs may display: stereotype bias and deviation bias. Stereotype bias refers to when LLMs consistently associate specific traits with a particular demographic group. Deviation bias reflects the disparity between the demographic distributions extracted from LLM-generated content and real-world demographic distributions. By asking four advanced LLMs to generate profiles of individuals, we examine the associations between each demographic group and attributes such as political affiliation, religion, and sexual orientation. Our experimental results show that all examined LLMs exhibit both significant stereotype bias and deviation bias towards multiple groups. Our findings uncover the biases that occur when LLMs infer user attributes and shed light on the potential harms of LLM-generated outputs.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 19:03:57 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 20:38:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Daniel",
""
],
[
"Brignac",
"Eli",
""
],
[
"Mao",
"Minjia",
""
],
[
"Fang",
"Xiao",
""
]
] |
2508.06697
|
Tomas Berggren
|
Tomas Berggren and Marianna Russkikh
|
Perfect t-embeddings and the octahedron equation of the two-periodic Aztec diamond
|
16 pages, 4 figures. Improved introduction, including a slight rephrasing of Theorem 1.1
| null | null | null |
math-ph math.CO math.MP math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper explores the connection between perfect t-embeddings and the octahedron equation in the setting of the two-periodic Aztec diamond. In particular, we show that the positions of both the t-embedding and the corresponding origami map can be expressed as sums of density functions arising from solutions to the octahedron equation with appropriate flat initial conditions.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 20:51:36 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 09:13:23 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Berggren",
"Tomas",
""
],
[
"Russkikh",
"Marianna",
""
]
] |
2508.06842
|
Seonggyu Lee
|
Seonggyu Lee, Sein Cheong, Sangwook Han, Kihyuk Kim, Jong Won Shin
|
Speech Enhancement based on cascaded two flows
|
Accepted at Interspeech 2025
| null | null | null |
eess.AS eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Speech enhancement (SE) based on diffusion probabilistic models has exhibited impressive performance, while requiring a relatively high number of function evaluations (NFE). Recently, SE based on flow matching has been proposed, which showed competitive performance with a small NFE. Early approaches adopted the noisy speech as the only conditioning variable. There have been other approaches which utilize speech enhanced with a predictive model as another conditioning variable and to sample an initial value, but they require a separate predictive model on top of the generative SE model. In this work, we propose to employ an identical model based on flow matching for both SE and generating enhanced speech used as an initial starting point and a conditioning variable. Experimental results showed that the proposed method required the same or fewer NFEs even with two cascaded generative methods while achieving equivalent or better performances to the previous baselines.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 05:50:46 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 12:25:15 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 11:09:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lee",
"Seonggyu",
""
],
[
"Cheong",
"Sein",
""
],
[
"Han",
"Sangwook",
""
],
[
"Kim",
"Kihyuk",
""
],
[
"Shin",
"Jong Won",
""
]
] |
2508.06961
|
Lasko Basnarkov
|
Lasko Basnarkov
|
Semi-Markov random walk on complex networks
| null | null | null | null |
physics.soc-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We present a semi-Markov model of random walk on complex networks in discrete and continuous-time scenario. In the general setting of the semi-Markov chains, the duration of stay at given node - the sojourn time - is random, and the probability to transition to a neighbor depends on the sojourn time. Analytical formulae for the average sojourn time and the node occupation probability of infinite walk are presented and verified with Monte Carlo simulations for two examples. We propose an application of the semi-Markovian random walk for ranking web pages determined by the fraction of time that infinite random surfer spends on a web page - time rank, as an alternative to the existing PageRank that relies on the fraction of visits - visit rank.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 12:16:12 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 08:58:59 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Basnarkov",
"Lasko",
""
]
] |
2508.07227
|
Siyuan He
|
Siyuan He, Zhantong Zhu, Yandong He, Tianyu Jia
|
LP-Spec: Leveraging LPDDR PIM for Efficient LLM Mobile Speculative Inference with Architecture-Dataflow Co-Optimization
|
there are some data inaccuracies in section III
| null | null | null |
cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
LLM inference on mobile devices faces extraneous challenges due to limited memory bandwidth and computational resources. To address these issues, speculative inference and processing-in-memory (PIM) techniques have been explored at the algorithmic and hardware levels. However, speculative inference results in more compute-intensive GEMM operations, creating new design trade-offs for existing GEMV-accelerated PIM architectures. Furthermore, there exists a significant amount of redundant draft tokens in tree-based speculative inference, necessitating efficient token management schemes to minimize energy consumption. In this work, we present LP-Spec, an architecture-dataflow co-design leveraging hybrid LPDDR5 performance-enhanced PIM architecture with draft token pruning and dynamic workload scheduling to accelerate LLM speculative inference. A near-data memory controller is proposed to enable data reallocation between DRAM and PIM banks. Furthermore, a data allocation unit based on the hardware-aware draft token pruner is developed to minimize energy consumption and fully exploit parallel execution opportunities. Compared to end-to-end LLM inference on other mobile solutions such as mobile NPUs or GEMV-accelerated PIMs, our LP-Spec achieves 13.21x, 7.56x, and 99.87x improvements in performance, energy efficiency, and energy-delay-product (EDP). Compared with prior AttAcc PIM and RTX 3090 GPU, LP-Spec can obtain 12.83x and 415.31x EDP reduction benefits.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 08:11:08 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 06:05:42 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"He",
"Siyuan",
""
],
[
"Zhu",
"Zhantong",
""
],
[
"He",
"Yandong",
""
],
[
"Jia",
"Tianyu",
""
]
] |
2508.07252
|
Siyuan He
|
Siyuan He, Peiran Yan, Yandong He, Youwei Zhuo, Tianyu Jia
|
Tasa: Thermal-aware 3D-Stacked Architecture Design with Bandwidth Sharing for LLM Inference
|
there are some data inaccuracies in section V
| null | null | null |
cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The autoregressive decoding in LLMs is the major inference bottleneck due to the memory-intensive operations and limited hardware bandwidth. 3D-stacked architecture is a promising solution with significantly improved memory bandwidth, which vertically stacked multi DRAM dies on top of logic die. However, our experiments also show the 3D-stacked architecture faces severer thermal issues compared to 2D architecture, in terms of thermal temperature, gradient and scalability. To better exploit the potential of 3D-stacked architecture, we present Tasa, a heterogeneous architecture with cross-stack thermal optimizations to balance the temperature distribution and maximize the performance under the thermal constraints. High-performance core is designed for compute-intensive operations, while high-efficiency core is used for memory-intensive operators, e.g. attention layers. Furthermore, we propose a bandwidth sharing scheduling to improve the bandwidth utilization in such heterogeneous architecture. Extensive thermal experiments show that our Tasa architecture demonstrates greater scalability compared with the homogeneous 3D-stacked architecture, i.e. up to 5.55 $\tccentigrade$, 9.37 $\tccentigrade$, and 7.91 $\tccentigrade$ peak temperature reduction for 48, 60, and 72 core configurations. Our experimental for Llama-65B and GPT-3 66B inferences also demonstrate 2.85x and 2.21x speedup are obtained over the GPU baselines and state-of-the-art heterogeneous PIM-based LLM accelerator
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 09:09:34 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 06:07:57 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"He",
"Siyuan",
""
],
[
"Yan",
"Peiran",
""
],
[
"He",
"Yandong",
""
],
[
"Zhuo",
"Youwei",
""
],
[
"Jia",
"Tianyu",
""
]
] |
2508.07266
|
Andrew Mansour
|
V. V. Flambaum and A. J. Mansour
|
Constraints on the Variation of the QCD Interaction Scale $\Lambda_{\text{QCD}}$
| null | null | null | null |
hep-ph astro-ph.CO nucl-th physics.atom-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Laboratory and astrophysical tests of ''constant variation'' have so far concentrated on the dimensionless fine-structure constant $\alpha$ and on the electron or quark mass ratios $X_{e,q}=m_{e,q}/\Lambda_{\text{QCD}}$, treating the QCD scale $\Lambda_{\text{QCD}}$ as unchangeable. Certain beyond Standard Model frameworks, most notably those with a dark matter or dark energy scalar field $\phi$ coupling with the gluon field, would make $\Lambda_{\text{QCD}}$ itself time dependent while leaving $\alpha$ and the electron mass untouched. Under the minimal assumption that this gluonic channel is the sole $\phi$ interaction, we recast state-of-the-art atomic clock comparisons into $\dot{\Lambda}_{\text{QCD}}/\Lambda_{\text{QCD}}=(3.2 \pm 3.5) \times 10^{-17} \ \text{yr}^{-1}$ limits, translate the isotope yields of the 1.8-Gyr-old Oklo natural reactor into a complementary geophysical limit of $|\delta\Lambda_{\text{QCD}}/\Lambda_{\text{QCD}}|<2\times10^{-9}$ over that time span, corresponding to the linear drift limit $|\dot{\Lambda}_{\text{QCD}}/\Lambda_{\text{QCD}}|<1\times10^{-18} \text{yr}^{-1}$, and show that the proposed $8.4$ eV $^{229}$Th nuclear clock would amplify a putative $\Lambda_{\text{QCD}}$ drift by four orders of magnitude compared with present atomic clocks. We also obtain constraints from quasar absorption spectra and Big Bang Nucleosynthesis data.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 09:39:33 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 02:57:18 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 08:26:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Flambaum",
"V. V.",
""
],
[
"Mansour",
"A. J.",
""
]
] |
2508.07425
|
Aliaksei Mikhaliuk
|
Jutta Kunz, Aliaksei Mikhaliuk, Yakov Shnir
|
Two types of boson stars in $U(1)$ gauged $3+1$-dimensional $O(3)$ sigma-model
|
18 pages, 8 figures
| null | null | null |
hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate boson stars in an $O(3)$ scalar field theory with a symmetry-breaking potential. By constructing numerically spherically symmetric solutions, we demonstrate that the model gives rise to a rich set of field configurations. The negative coupling constant of the scalar self-interactions allows for two types of boson stars: Type I solutions represent the usual boson stars, that emerge from the vacuum as the boson frequency is decreased below the boson mass, whereas type II boson stars emerge from a set of static soliton solutions. Depending on the strengths of the gravitational and the electric coupling constants, both types or only one type is present. At a critical set of coupling constants, both types undergo a bifurcation. There the spirals of both types disconnect from their branches and reconnect with each other, while the remaining branches of both types also reconnect with each other.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 16:54:48 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 18:06:57 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Kunz",
"Jutta",
""
],
[
"Mikhaliuk",
"Aliaksei",
""
],
[
"Shnir",
"Yakov",
""
]
] |
2508.07511
|
Raj Dahya
|
Raj Dahya
|
Dilations of non-Markovian dynamical systems on graphs
|
36 pages; corrected and simplified construction steps in the Ist and IInd cover extension lemmata; corrected minor typos
| null | null | null |
math.FA math-ph math.GR math.MP math.OA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
To generalise evolution families we consider systems of contractions $\{\varphi(u, v)\}_{(u, v) \in E}$ defined on the edges of a graph $\mathcal{G} = (\Omega, E)$. In this setup the Markov property, or \emph{divisibility}, can be modelled via $\varphi(u, v)\varphi(v, w) = \varphi(u, w)$ for edges $(u, v), (v, w), (u, w) \in E$. We obtain results in three settings: 1) contractive Banach space operators; 2) positive unital maps on $C^{\ast}$-algebras; and 3) CPTP-maps on trace class operators on a Hilbert space. In the discrete setting, we are able to dilate possibly indivisible families of contractions to divisible families of operators with 'nice' properties (viz. surjective isometries resp. $C^{\ast}$-algebraic automorphisms resp. unitary representations). In the special case of linearly ordered graphs equipped with the order topology, we establish sufficient conditions for strongly continuous dilations of possibly indivisible families in the Banach space and $C^{\ast}$-algebra contexts. To achieve these results we work with string-rewriting systems, and make use of and extend dilation theorems of Stroescu [44], Kraus [23, 24], and vom Ende--Dirr [50].
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 23:54:50 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 13:03:58 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 14:29:13 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dahya",
"Raj",
""
]
] |
2508.07571
|
Xingwu Chen
|
Xingwu Chen, Miao Lu, Beining Wu, Difan Zou
|
Towards Theoretical Understanding of Transformer Test-Time Computing: Investigation on In-Context Linear Regression
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Using more test-time computation during language model inference, such as generating more intermediate thoughts or sampling multiple candidate answers, has proven effective in significantly improving model performance. This paper takes an initial step toward bridging the gap between practical language model inference and theoretical transformer analysis by incorporating randomness and sampling. We focus on in-context linear regression with continuous/binary coefficients, where our framework simulates language model decoding through noise injection and binary coefficient sampling. Through this framework, we provide detailed analyses of widely adopted inference techniques. Supported by empirical results, our theoretical framework and analysis demonstrate the potential for offering new insights into understanding inference behaviors in real-world language models.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 03:05:36 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:54:07 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chen",
"Xingwu",
""
],
[
"Lu",
"Miao",
""
],
[
"Wu",
"Beining",
""
],
[
"Zou",
"Difan",
""
]
] |
2508.07623
|
Zhun Gou
|
Zhun Gou, Nan-Jing Huang, Jian-Hao Kang, Jen-Chih Yao
|
Nash Equilibria of Noncooperative/Mixed Differential Games with Density Constraints in Infinite Dimensions
| null | null | null | null |
math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
Motivated by Cournot models, this paper proposes novel models of the noncooperative and cooperative differential games with density constraints in infinite dimensions, where markets consist of infinite firms and demand dynamics are governed by controlled differential equations. Markets engage in noncooperative competition with each other, while firms within each market engage in noncooperative or cooperative games. The main problems are to find the noncooperative Nash equilibrium (NNE) of the noncooperative differential game and the mixed Nash equilibrium (MNE) of the mixed noncooperative and cooperative differential game. Moreover, fundamental relationship is established between noncooperative/mixed differential game with density constraints and infinite-dimensional differential variational inequalities with density constraints. By variational analysis, it is proved under two conditions with certain symmetry that both of the two equilibrium problems can be reduced to solving systems of finite-dimensional projection equations with integral constraints by iterative computational methods. Crucially, the two conditions with certain symmetry, ensuring the uniqueness of the NNE and the MNE, provide theoretical foundations for strategic decision making regarding competitive versus cooperative market behaviors. Finally, the theoretical framework is validated through numerical simulations demonstrating the efficacy of our results.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 05:06:03 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:38:22 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Gou",
"Zhun",
""
],
[
"Huang",
"Nan-Jing",
""
],
[
"Kang",
"Jian-Hao",
""
],
[
"Yao",
"Jen-Chih",
""
]
] |
2508.07627
|
H Chan
|
H Chan
|
Nonlinear Systems in Wireless Power Transfer Applications
| null | null | null | null |
eess.SY cs.SY
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
As a novel pattern of energization, the wireless power transfer (WPT) offers a brand-new way to the energy acquisition for electric-driven devices, thus alleviating the over-dependence on the battery. This report presents three types of WPT systems that use nonlinear control methods, in order to acquire an in-depth understanding of the course of Nonlinear Systems.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 05:15:11 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 03:04:40 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 15:13:39 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chan",
"H",
""
]
] |
2508.07729
|
Juliusz Banecki
|
Juliusz Banecki
|
Uniform rationality of products with the plane
| null | null | null | null |
math.AG
|
http://creativecommons.org/licenses/by/4.0/
|
Let $X$ be a nonsingular rational variety. We prove that $X\times \mathbb{C}^2$ is uniformly rational. It follows that nonsingular stably rational varieties are stably uniformly rational.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 08:01:14 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 19:30:21 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Banecki",
"Juliusz",
""
]
] |
2508.07734
|
Shenghao Hua
|
Shenghao Hua
|
Decorrelation phenomena in mixed moments for orthogonal families of $L$-Functions and their applications: automorphic periods, Fourier coefficients of half-integral weight forms, and isotropy subgroups of Tate--Shafarevich groups
|
15 pages. Comments welcome
| null | null | null |
math.NT
|
http://creativecommons.org/licenses/by/4.0/
|
The Keating--Snaith conjecture for orthogonal families may be viewed as analogous to a Gaussian distribution with a negative mean, and the possibility that mixed moments resemble a composition of independent moments, these two insights were combined and applied in Lester and Radziwi\l\l's proof of quantum unique ergodicity for half-integral weight automorphic forms, via Soundararajan's method under the Generalized Riemann Hypothesis (GRH). This observation also yields a crucial and nontrivial saving in the resolution of certain arithmetic problems.
Inspired by this, we select a series of typical mixed orthogonal families of $L$-functions: $\mathrm{GL}_2$ quadratic twisted families. Under the assumptions of the GRH and the Generalized Ramanujan Conjecture, we prove upper bound estimates for their moments and present the following three arithmetic applications:
i) The decorrelation of automorphic periods averaged over imaginary quadratic fields.
ii) The decorrelation of Fourier coefficients of half-integral weight modular forms.
iii) The decorrelation of the analytic orders of isotropy subgroups of Tate--Shafarevich groups of elliptic curves under quadratic twists.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 08:12:24 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 16:10:37 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hua",
"Shenghao",
""
]
] |
2508.07870
|
Mohammad H. Ansari
|
Julian Rapp, Radhika H. Joshi, Alwin van Steensel, Yuli V. Nazarov and Mohammad H. Ansari
|
Information Transport in Classical-Quantum Hybrid System
|
16 pages, 11 figures, References updated
| null | null | null |
quant-ph cond-mat.mes-hall gr-qc hep-th math-ph math.MP
|
http://creativecommons.org/licenses/by/4.0/
|
Many important quantities in quantum information science, such as entropy and entanglement, are non-linear functions of the density matrix and cannot be expressed as operator observables. Standard open-system approaches evolve only a single copy of the density matrix, making it impossible to track the dynamics of such quantities. A formalism proposed by some of the present authors addressed this challenge by evolving multiple virtual replicas, but was limited to the weak-coupling regime. Here, we extend this approach to strong coupling between a quantum system and classical environments. The resulting multi-replica master equation enables direct evaluation of entropy flow and related metrics in strongly hybridized quantum-classical systems. Our results show that quantum coherence and hybridization jointly suppress net entropy transfer, creating a thermodynamic bottleneck. This framework provides a general tool for studying entropy dynamics and guiding the design of more robust, resource-efficient quantum hardware.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 11:41:03 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 21:01:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Rapp",
"Julian",
""
],
[
"Joshi",
"Radhika H.",
""
],
[
"van Steensel",
"Alwin",
""
],
[
"Nazarov",
"Yuli V.",
""
],
[
"Ansari",
"Mohammad H.",
""
]
] |
2508.07917
|
Jiafei Duan
|
Jason Lee, Jiafei Duan, Haoquan Fang, Yuquan Deng, Shuo Liu, Boyang Li, Bohan Fang, Jieyu Zhang, Yi Ru Wang, Sangho Lee, Winson Han, Wilbert Pumacay, Angelica Wu, Rose Hendrix, Karen Farley, Eli VanderBilt, Ali Farhadi, Dieter Fox, Ranjay Krishna
|
MolmoAct: Action Reasoning Models that can Reason in Space
|
Appendix include. Code, Data and Weights: https://allenai.org/blog/molmoact
| null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reasoning is central to purposeful action, yet most robotic foundation models map perception and instructions directly to control, which limits adaptability, generalization, and semantic grounding. We introduce Action Reasoning Models (ARMs), a class of robotic foundation models that integrate perception, planning, and control through a structured three-stage pipeline. Our model, MolmoAct, encodes observations and instructions into depth-aware perception tokens, generates mid-level spatial plans as editable trajectory traces, and predicts precise low-level actions, enabling explainable and steerable behavior. MolmoAct-7B-D achieves strong performance across simulation and real-world settings: 70.5% zero-shot accuracy on SimplerEnv Visual Matching tasks, surpassing closed-source Pi-0 and GR00T N1; 86.6% average success on LIBERO, including an additional 6.3% gain over ThinkAct on long-horizon tasks; and in real-world fine-tuning, an additional 10% (single-arm) and an additional 22.7% (bimanual) task progression over Pi-0-FAST. It also outperforms baselines by an additional 23.3% on out-of-distribution generalization and achieves top human-preference scores for open-ended instruction following and trajectory steering. Furthermore, we release, for the first time, the MolmoAct Dataset -- a mid-training robot dataset comprising over 10,000 high quality robot trajectories across diverse scenarios and tasks. Training with this dataset yields an average 5.5% improvement in general performance over the base model. We release all model weights, training code, our collected dataset, and our action reasoning dataset, establishing MolmoAct as both a state-of-the-art robotics foundation model and an open blueprint for building ARMs that transform perception into purposeful action through structured reasoning. Blogpost: https://allenai.org/blog/molmoact
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 12:32:45 GMT"
},
{
"version": "v2",
"created": "Tue, 12 Aug 2025 19:00:21 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 04:01:49 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lee",
"Jason",
""
],
[
"Duan",
"Jiafei",
""
],
[
"Fang",
"Haoquan",
""
],
[
"Deng",
"Yuquan",
""
],
[
"Liu",
"Shuo",
""
],
[
"Li",
"Boyang",
""
],
[
"Fang",
"Bohan",
""
],
[
"Zhang",
"Jieyu",
""
],
[
"Wang",
"Yi Ru",
""
],
[
"Lee",
"Sangho",
""
],
[
"Han",
"Winson",
""
],
[
"Pumacay",
"Wilbert",
""
],
[
"Wu",
"Angelica",
""
],
[
"Hendrix",
"Rose",
""
],
[
"Farley",
"Karen",
""
],
[
"VanderBilt",
"Eli",
""
],
[
"Farhadi",
"Ali",
""
],
[
"Fox",
"Dieter",
""
],
[
"Krishna",
"Ranjay",
""
]
] |
2508.08025
|
Marius Huber
|
Marius Huber and Patrick Schnider
|
Flagifying the Dowker Complex
|
14 pages, comments welcome; fixed typos
| null | null | null |
math.AT cs.CG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Dowker complex $\mathrm{D}_{R}(X,Y)$ is a simplicial complex capturing the topological interplay between two finite sets $X$ and $Y$ under some relation $R\subseteq X\times Y$. While its definition is asymmetric, the famous Dowker duality states that $\mathrm{D}_{R}(X,Y)$ and $\mathrm{D}_{R}(Y,X)$ have homotopy equivalent geometric realizations. We introduce the Dowker-Rips complex $\mathrm{DR}_{R}(X,Y)$, defined as the flagification of the Dowker complex or, equivalently, as the maximal simplicial complex whose $1$-skeleton coincides with that of $\mathrm{D}_{R}(X,Y)$. This is motivated by applications in topological data analysis, since as a flag complex, the Dowker-Rips complex is less expensive to compute than the Dowker complex. While the Dowker duality does not hold for Dowker-Rips complexes in general, we show that one still has that $\mathrm{H}_{i}(\mathrm{DR}_{R}(X,Y))\cong\mathrm{H}_{i}(\mathrm{DR}_{R}(Y,X))$ for $i=0,1$. We further show that this weakened duality extends to the setting of persistent homology, and quantify the ``failure" of the Dowker duality in homological dimensions higher than $1$ by means of interleavings. This makes the Dowker-Rips complex a less expensive, approximate version of the Dowker complex that is usable in topological data analysis. Indeed, we provide a Python implementation of the Dowker-Rips complex and, as an application, we show that it can be used as a drop-in replacement for the Dowker complex in a tumor microenvironment classification pipeline. In that pipeline, using the Dowker-Rips complex leads to increase in speed while retaining classification performance.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 14:30:17 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 14:06:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Huber",
"Marius",
""
],
[
"Schnider",
"Patrick",
""
]
] |
2508.08639
|
Yoshiki Fukusumi
|
Yoshiki Fukusumi, Shinichiro Yahagi
|
Extending fusion rules with finite subgroups: For a general understanding of quotient or gauging
|
2 figures. Typos are corrected and references are added (v2)
| null | null | null |
hep-th cond-mat.str-el math-ph math.MP math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce a general method for realizing simple current extensions of the conformal field theories. We systematically obtain the $Z_{N}$ symmetry extended fusion ring of bulk and chiral theories and the corresponding modular partition functions with nonanomalous subgroup $Z_{n}(\subset Z_{N})$. The bulk (or nonchiral) fusion ring provides fundamental algebraic data for conformal bootstrap, and the chiral fusion ring provides fundamental data for the graded symmetry topological field theories. The latter also provides algebraic data of smeared boundary conformal field theories describing zero modes of the extended models. For more general multicomponent or coupled systems, we also obtain a new series of extended theories. By applying the folding trick to the resultant coupled theories, their partition functions correspond to charged or gapped domain walls or massless renormalization group flows preserving quotient group structures. This work opens new research directions in studying the classification of conformal field theories and the corresponding topological quantum field theories (or topological orders) by establishing the traditional methods in abstract algebra and modular form.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 05:05:35 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 23:32:59 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fukusumi",
"Yoshiki",
""
],
[
"Yahagi",
"Shinichiro",
""
]
] |
2508.09088
|
Swamynadhan Madras Jayasankar
|
M. J. Swamynadhan, Gwan Yeong Jung, Pravan Omprakash, Rohan Mishra
|
Design Rules and Discovery of Face-Sharing Hexagonal Perovskites
| null | null | null | null |
cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
Hexagonal perovskites with face-sharing octahedral connectivity are an underexplored class of materials. We propose quantitative design principles for stabilizing face-sharing ABX3 hexagonal perovskites based on a comparative analysis of oxides and sulfides. By mapping structural preferences across a phase-space defined by an electronegativity-corrected tolerance factor and the Shannon radius of the A-site cations, we identify distinct thresholds that separate hexagonal phases from competing cubic polymorphs having corner-sharing octahedral connectivity. Our analysis reveals that sulfides differ significantly from oxides due to the increased covalency of the transition metal-sulfur bonds, which enables broader compositional flexibility. Applying these principles, we predict a set of thermodynamically formable ABO3 and ABS3 compounds that are likely to adopt face-sharing octahedral connectivity. These findings establish a predictive framework for designing hexagonal perovskites, highlighting sulfides as promising candidates for obtaining quasi-one-dimensional materials having transition-metal cations for novel ferroic phenomena.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 17:11:46 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 01:52:47 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Swamynadhan",
"M. J.",
""
],
[
"Jung",
"Gwan Yeong",
""
],
[
"Omprakash",
"Pravan",
""
],
[
"Mishra",
"Rohan",
""
]
] |
2508.09146
|
Shugang Hao
|
Shugang Hao, Hongbo Li and Lingjie Duan
|
To Theoretically Understand Transformer-Based In-Context Learning for Optimizing CSMA
| null | null | null | null |
cs.LG cs.AI cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The binary exponential backoff scheme is widely used in WiFi 7 and still incurs poor throughput performance under dynamic channel environments. Recent model-based approaches (e.g., non-persistent and $p$-persistent CSMA) simply optimize backoff strategies under a known and fixed node density, still leading to a large throughput loss due to inaccurate node density estimation. This paper is the first to propose LLM transformer-based in-context learning (ICL) theory for optimizing channel access. We design a transformer-based ICL optimizer to pre-collect collision-threshold data examples and a query collision case. They are constructed as a prompt as the input for the transformer to learn the pattern, which then generates a predicted contention window threshold (CWT). To train the transformer for effective ICL, we develop an efficient algorithm and guarantee a near-optimal CWT prediction within limited training steps. As it may be hard to gather perfect data examples for ICL in practice, we further extend to allow erroneous data input in the prompt. We prove that our optimizer maintains minimal prediction and throughput deviations from the optimal values. Experimental results on NS-3 further demonstrate our approach's fast convergence and near-optimal throughput over existing model-based and DRL-based approaches under unknown node densities.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 23:31:23 GMT"
},
{
"version": "v2",
"created": "Thu, 14 Aug 2025 01:13:56 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 01:29:34 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hao",
"Shugang",
""
],
[
"Li",
"Hongbo",
""
],
[
"Duan",
"Lingjie",
""
]
] |
2508.09190
|
Bing Han
|
Bing Han, Feifei Zhao, Dongcheng Zhao, Guobin Shen, Ping Wu, Yu Shi, Yi Zeng
|
Fine-Grained Safety Neurons with Training-Free Continual Projection to Reduce LLM Fine Tuning Risks
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Fine-tuning as service injects domain-specific knowledge into large language models (LLMs), while challenging the original alignment mechanisms and introducing safety risks. A series of defense strategies have been proposed for the alignment, fine-tuning, and post-fine-tuning phases, where most post-fine-tuning defenses rely on coarse-grained safety layer mapping. These methods lack a comprehensive consideration of both safety layers and fine-grained neurons, limiting their ability to efficiently balance safety and utility. To address this, we propose the Fine-Grained Safety Neurons (FGSN) with Training-Free Continual Projection method to reduce the fine-tuning safety risks. FGSN inherently integrates the multi-scale interactions between safety layers and neurons, localizing sparser and more precise fine-grained safety neurons while minimizing interference with downstream task neurons. We then project the safety neuron parameters onto safety directions, improving model safety while aligning more closely with human preferences. Extensive experiments across multiple fine-tuned LLM models demonstrate that our method significantly reduce harmfulness scores and attack success rates with minimal parameter modifications, while preserving the model's utility. Furthermore, by introducing a task-specific, multi-dimensional heterogeneous safety neuron cluster optimization mechanism, we achieve continual defense and generalization capability against unforeseen emerging safety concerns.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 03:20:25 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 09:06:26 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Han",
"Bing",
""
],
[
"Zhao",
"Feifei",
""
],
[
"Zhao",
"Dongcheng",
""
],
[
"Shen",
"Guobin",
""
],
[
"Wu",
"Ping",
""
],
[
"Shi",
"Yu",
""
],
[
"Zeng",
"Yi",
""
]
] |
2508.09288
|
Aayush Gupta
|
Aayush Gupta
|
Can AI Keep a Secret? Contextual Integrity Verification: A Provable Security Architecture for LLMs
|
2 figures, 3 tables; code and certification harness: https://github.com/ayushgupta4897/Contextual-Integrity-Verification ; Elite-Attack dataset: https://huggingface.co/datasets/zyushg/elite-attack
| null | null | null |
cs.CR cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Large language models (LLMs) remain acutely vulnerable to prompt injection and related jailbreak attacks; heuristic guardrails (rules, filters, LLM judges) are routinely bypassed. We present Contextual Integrity Verification (CIV), an inference-time security architecture that attaches cryptographically signed provenance labels to every token and enforces a source-trust lattice inside the transformer via a pre-softmax hard attention mask (with optional FFN/residual gating). CIV provides deterministic, per-token non-interference guarantees on frozen models: lower-trust tokens cannot influence higher-trust representations. On benchmarks derived from recent taxonomies of prompt-injection vectors (Elite-Attack + SoK-246), CIV attains 0% attack success rate under the stated threat model while preserving 93.1% token-level similarity and showing no degradation in model perplexity on benign tasks; we note a latency overhead attributable to a non-optimized data path. Because CIV is a lightweight patch -- no fine-tuning required -- we demonstrate drop-in protection for Llama-3-8B and Mistral-7B. We release a reference implementation, an automated certification harness, and the Elite-Attack corpus to support reproducible research.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 18:47:30 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 18:20:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Gupta",
"Aayush",
""
]
] |
2508.09539
|
Yongqi Fan
|
Yongqi Fan, Xiaoyang Chen, Dezhi Ye, Jie Liu, Haijin Liang, Jin Ma, Ben He, Yingfei Sun, Tong Ruan
|
TFRank: Think-Free Reasoning Enables Practical Pointwise LLM Ranking
| null | null | null | null |
cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reasoning-intensive ranking models built on Large Language Models (LLMs) have made notable progress, but existing approaches often rely on large-scale LLMs and explicit Chain-of-Thought (CoT) reasoning, resulting in high computational cost and latency that limit real-world use. To address this, we propose \textbf{TFRank}, an efficient pointwise reasoning ranker based on small-scale LLMs. To improve ranking performance, TFRank effectively integrates CoT data, fine-grained score supervision, and multi-task training. Furthermore, it achieves an efficient ``\textbf{T}hink-\textbf{F}ree" reasoning capability by employing a ``think-mode switch'' and pointwise format constraints. Specifically, this allows the model to leverage explicit reasoning during training while delivering precise relevance scores for complex queries at inference without generating any reasoning chains. Experiments show that TFRank (e.g., 1.7B) achieves performance comparable to models with four times more parameters on the BRIGHT benchmark, and demonstrates strong competitiveness on the BEIR benchmark. Further analysis shows that TFRank achieves an effective balance between performance and efficiency, providing a practical solution for integrating advanced reasoning into real-world systems. Our code and data are released in the repository: https://github.com/JOHNNY-fans/TFRank.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 06:47:58 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 04:21:43 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fan",
"Yongqi",
""
],
[
"Chen",
"Xiaoyang",
""
],
[
"Ye",
"Dezhi",
""
],
[
"Liu",
"Jie",
""
],
[
"Liang",
"Haijin",
""
],
[
"Ma",
"Jin",
""
],
[
"He",
"Ben",
""
],
[
"Sun",
"Yingfei",
""
],
[
"Ruan",
"Tong",
""
]
] |
2508.09872
|
Harold Erazo
|
Harold Erazo
|
New examples of $M\setminus L$: intruder sets
|
44 pages, 4 tables, 3 figures
| null | null | null |
math.NT math.CO
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We exhibit new examples of regions of $M\setminus L$ where $M$ and $L$ denote the Markov and Lagrange spectra, respectively. These regions have a different nature from all known regions studied so far: they contain \emph{intruder sets} associated with distinct combinatorics that trespass the region where self-replication holds. Our construction follows the usual self-replication method but replaces the standard local uniqueness condition with a more flexible and weaker property. These examples emerged from a large-scale computational search for regions of $M\setminus L$, which indicates that many such regions with intruder sets exist. We conclude with some open problems about these new regions.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 15:11:37 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 14:04:00 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Erazo",
"Harold",
""
]
] |
2508.09986
|
Farvah Mahmoudi
|
T. Hurth, F. Mahmoudi, Y. Monceaux and S. Neshatpour
|
Data-driven analyses and model-independent fits for present $b\to s \ell \ell$ results
|
29 pages, 11 figures, 9 tables
| null | null |
CERN-TH-2025-164, MITP/25-055
|
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a critical assessment of the present $B$ anomalies in the exclusive $b \to s \ell\ell$ mode based on the QCD factorisation approach (QCDf). In particular, we analyse the impact of different local form factor calculations and of the largest bin in the low-$q^2$ region.
We also present a model-independent analysis of the new results of the CMS experiment on the $B \to K^* \mu^+\mu^-$ angular observables and compare them with the corresponding LHCb data. In addition, we update the global fit by including all $b \to s$ observables incorporating the new data from CMS. In these analyses, we use 10% or higher guesstimates of the non-factorisable power corrections as additional uncertainties, serving as a placeholder for robust estimates of these contributions.
Updating earlier results, we also analyse the combined LHCb and CMS data on the $B \to K^* \mu^+\mu^-$ angular observables using data-driven approaches to find indications whether these tensions between the QCDf predictions and the present data are due to underestimated subleading hadronic contributions or due to new physics effects.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 17:59:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hurth",
"T.",
""
],
[
"Mahmoudi",
"F.",
""
],
[
"Monceaux",
"Y.",
""
],
[
"Neshatpour",
"S.",
""
]
] |
2508.10040
|
V\'itor Louren\c{c}o
|
V\'itor N. Louren\c{c}o and Aline Paes and Tillman Weyde
|
Exploring Content and Social Connections of Fake News with Explainable Text and Graph Learning
|
Accepted to publication at the 35th Brazilian Conference on Intelligent Systems, BRACIS 2025. -- This submitted manuscript has not undergone any post-submission improvements or corrections. The Version of Record of this contribution will be provided when available
| null | null | null |
cs.SI cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
The global spread of misinformation and concerns about content trustworthiness have driven the development of automated fact-checking systems. Since false information often exploits social media dynamics such as "likes" and user networks to amplify its reach, effective solutions must go beyond content analysis to incorporate these factors. Moreover, simply labelling content as false can be ineffective or even reinforce biases such as automation and confirmation bias. This paper proposes an explainable framework that combines content, social media, and graph-based features to enhance fact-checking. It integrates a misinformation classifier with explainability techniques to deliver complete and interpretable insights supporting classification decisions. Experiments demonstrate that multimodal information improves performance over single modalities, with evaluations conducted on datasets in English, Spanish, and Portuguese. Additionally, the framework's explanations were assessed for interpretability, trustworthiness, and robustness with a novel protocol, showing that it effectively generates human-understandable justifications for its predictions.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 12:03:37 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 11:35:59 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 10:32:47 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lourenço",
"Vítor N.",
""
],
[
"Paes",
"Aline",
""
],
[
"Weyde",
"Tillman",
""
]
] |
2508.10131
|
Tejinder P. Singh
|
Tejinder P. Singh
|
Fermion mass ratios from the exceptional Jordan algebra
|
v2: 123 pages, 2 figures. (1) Added Sec. II (physical motivation for using J_3(O_C) and associated dynamics) (2) Added an E_6 covariant Lagrangian and UV anchors, in Sec. III (3) Added Fig. 2 on p.48 showing Sym^3((\bf 3}) as an equilateral triangle (4) Added Appendix G (Roadmap, explains in detail the Dynkin swap, Clebsch (2:1:1)). (5) updated refs, fixed minor typos. No changes to core results
| null | null | null |
hep-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We revisit the "three generations" problem and the pattern of charged-fermion masses from the vantage of octonionic and Clifford algebra structures. Working with the exceptional Jordan algebra J3(OC) (right-handed flavor) and the symmetric cube of SU(3) (left-handed charge frame), we show that a single minimal ladder in the symmetric cube, together with the Dynkin Z2 swap (the A2 diagram flip), leads to closed-form expressions for the square-root mass ratios of all three charged families. The universal Jordan spectrum (q - delta, q, q + delta) with a theoretically derived delta squared = 3/8 fixes the endpoint contrasts; fixed Clebsch factors (2, 1, 1) ensure rung cancellation ("edge universality") so that adjacent ratios depend only on which edge is taken. The down ladder determines one step, its Dynkin reflection gives the lepton ladder, and choosing the other outward leg from the middle yields the up sector.
From the same inputs we obtain compact CKM "root-sum rules": with one 1-2 phase and a mild 2-3 cross-family normalization, the framework reproduces the Cabibbo angle and Vcb and provides leading predictions for Vub and Vtd/Vts. We perform apples-to-apples phenomenology (common scheme/scale) and find consistency with current determinations within quoted uncertainties. Conceptually, rank-1 idempotents (points of the octonionic projective plane), fixed symmetric-cube Clebsches, and the Dynkin swap together account for why electric charge is generation-blind while masses follow the observed hierarchies, and they furnish clear, falsifiable mass-ratio relations beyond the Standard Model.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 18:53:19 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 17:08:33 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Singh",
"Tejinder P.",
""
]
] |
2508.10210
|
Ritesh Kumar
|
Rahul Jana, Shubham Dixit, Mrityunjay Sharma, Ritesh Kumar
|
An Explainable AI based approach for Monitoring Animal Health
| null | null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Monitoring cattle health and optimizing yield are key challenges faced by dairy farmers due to difficulties in tracking all animals on the farm. This work aims to showcase modern data-driven farming practices based on explainable machine learning(ML) methods that explain the activity and behaviour of dairy cattle (cows). Continuous data collection of 3-axis accelerometer sensors and usage of robust ML methodologies and algorithms, provide farmers and researchers with actionable information on cattle activity, allowing farmers to make informed decisions and incorporate sustainable practices. This study utilizes Bluetooth-based Internet of Things (IoT) devices and 4G networks for seamless data transmission, immediate analysis, inference generation, and explains the models performance with explainability frameworks. Special emphasis is put on the pre-processing of the accelerometers time series data, including the extraction of statistical characteristics, signal processing techniques, and lag-based features using the sliding window technique. Various hyperparameter-optimized ML models are evaluated across varying window lengths for activity classification. The k-nearest neighbour Classifier achieved the best performance, with AUC of mean 0.98 and standard deviation of 0.0026 on the training set and 0.99 on testing set). In order to ensure transparency, Explainable AI based frameworks such as SHAP is used to interpret feature importance that can be understood and used by practitioners. A detailed comparison of the important features, along with the stability analysis of selected features, supports development of explainable and practical ML models for sustainable livestock management.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 21:40:35 GMT"
},
{
"version": "v2",
"created": "Fri, 15 Aug 2025 08:12:18 GMT"
},
{
"version": "v3",
"created": "Mon, 18 Aug 2025 19:26:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Jana",
"Rahul",
""
],
[
"Dixit",
"Shubham",
""
],
[
"Sharma",
"Mrityunjay",
""
],
[
"Kumar",
"Ritesh",
""
]
] |
2508.10258
|
Allan Freitas
|
Allan Freitas, Benedito Leandro, Ernani Ribeiro Jr, and Guilherme Sabo
|
Geometric inequalities for electrostatic systems with boundary
|
Fixed typos. Comments are welcome!
| null | null | null |
math.DG
|
http://creativecommons.org/licenses/by/4.0/
|
In this article, we investigate electrostatic systems with a nonzero cosmological constant on compact manifolds with boundary. We establish new geometric properties for electrostatic manifolds in higher dimensions, extending previous results in the literature. Moreover, we prove sharp boundary estimates and isoperimetric-type inequalities for electrostatic manifolds, as well as volume and boundary inequalities involving the Brown-York and Hawking masses.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 00:56:00 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 02:46:57 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Freitas",
"Allan",
""
],
[
"Leandro",
"Benedito",
""
],
[
"Ribeiro",
"Ernani",
"Jr"
],
[
"Sabo",
"Guilherme",
""
]
] |
2508.10451
|
Shibo Fang
|
Shibo Fang, Zongmeng Yang, Jianhua Wang, Xingyue Yang, Jing Lu, Xiaotian Wang, Yee Sin Ang
|
Edgetronics in 2D Altermagnet via Real-Space-Spin Coupling
|
7 pages, 5 figures
| null | null | null |
cond-mat.mes-hall cond-mat.mtrl-sci
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The coupling between real-space coordinates and spin (r-s) provides an alternative route to achieve efficient spin manipulation in spintronics beyond the conventional momentum-spin (k-s) coupling paradigm. Here we demonstrate an unexpected manifestation of one-dimensional (1D) r-s coupling in two-dimensional (2D) altermagnetic second-order topological insulators, where the spin-split floating edge states - energetically isolated within the bulk band gap - emerge and exhibit both Neel-vector-dependent and electrically tunable behaviors. The 1D edge-spin r-s coupling ensures carrier transport to be exclusively carried by the edge states with quantized spin conductance, giving rise to an unconventional edge tunnel magnetoresistance (edge-TMR) effect that can be switched On or Off. As a proof of concept, we computationally design an edge-TMR device based on Cr2Se2O monolayer to demonstrate its edge transportation and controllability via the Neel order or electric field. Our findings propose a general prototype altermagnetic device for next-generation low-dimensional spintronics.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 08:39:38 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 05:10:55 GMT"
},
{
"version": "v3",
"created": "Tue, 19 Aug 2025 01:39:13 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fang",
"Shibo",
""
],
[
"Yang",
"Zongmeng",
""
],
[
"Wang",
"Jianhua",
""
],
[
"Yang",
"Xingyue",
""
],
[
"Lu",
"Jing",
""
],
[
"Wang",
"Xiaotian",
""
],
[
"Ang",
"Yee Sin",
""
]
] |
2508.10594
|
Yunfeng Zhao
|
Yunfeng Zhao, Yixin Liu, Shiyuan Li, Qingfeng Chen, Yu Zheng and Shirui Pan
|
FreeGAD: A Training-Free yet Effective Approach for Graph Anomaly Detection
|
Accepted by ClKM 2025
| null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Graph Anomaly Detection (GAD) aims to identify nodes that deviate from the majority within a graph, playing a crucial role in applications such as social networks and e-commerce. Despite the current advancements in deep learning-based GAD, existing approaches often suffer from high deployment costs and poor scalability due to their complex and resource-intensive training processes. Surprisingly, our empirical findings suggest that the training phase of deep GAD methods, commonly perceived as crucial, may actually contribute less to anomaly detection performance than expected. Inspired by this, we propose FreeGAD, a novel training-free yet effective GAD method. Specifically, it leverages an affinity-gated residual encoder to generate anomaly-aware representations. Meanwhile, FreeGAD identifies anchor nodes as pseudo-normal and anomalous guides, followed by calculating anomaly scores through anchor-guided statistical deviations. Extensive experiments demonstrate that FreeGAD achieves superior anomaly detection performance, efficiency, and scalability on multiple benchmark datasets from diverse domains, without any training or iterative optimization.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 12:37:20 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 01:47:34 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhao",
"Yunfeng",
""
],
[
"Liu",
"Yixin",
""
],
[
"Li",
"Shiyuan",
""
],
[
"Chen",
"Qingfeng",
""
],
[
"Zheng",
"Yu",
""
],
[
"Pan",
"Shirui",
""
]
] |
2508.10747
|
Sangwoo Jeon
|
Sangwoo Jeon, Juchul Shin, Gyeong-Tae Kim, YeonJe Cho and Seongwoo Kim
|
Scaling Up without Fading Out: Goal-Aware Sparse GNN for RL-based Generalized Planning
| null | null | null | null |
cs.AI cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Generalized planning using deep reinforcement learning (RL) combined with graph neural networks (GNNs) has shown promising results in various symbolic planning domains described by PDDL. However, existing approaches typically represent planning states as fully connected graphs, leading to a combinatorial explosion in edge information and substantial sparsity as problem scales grow, especially evident in large grid-based environments. This dense representation results in diluted node-level information, exponentially increases memory requirements, and ultimately makes learning infeasible for larger-scale problems. To address these challenges, we propose a sparse, goal-aware GNN representation that selectively encodes relevant local relationships and explicitly integrates spatial features related to the goal. We validate our approach by designing novel drone mission scenarios based on PDDL within a grid world, effectively simulating realistic mission execution environments. Our experimental results demonstrate that our method scales effectively to larger grid sizes previously infeasible with dense graph representations and substantially improves policy generalization and success rates. Our findings provide a practical foundation for addressing realistic, large-scale generalized planning tasks.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 15:30:28 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 12:06:36 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Jeon",
"Sangwoo",
""
],
[
"Shin",
"Juchul",
""
],
[
"Kim",
"Gyeong-Tae",
""
],
[
"Cho",
"YeonJe",
""
],
[
"Kim",
"Seongwoo",
""
]
] |
2508.10766
|
Jinlu Li
|
Jinlu Li
|
Frechet and Mordukhovich Derivative (Coderivative) and Covering Constant for Single-Valued Mapping in Euclidean Space with Applications (I)
|
This is a new research article
| null | null | null |
math.FA
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we study Frechet derivatives and Mordukhovich derivatives (or coderivatives) of single-valued mappings in Euclidean spaces. At first, we prove the guideline for calculating the Frechet derivatives of single-valued mappings by their partial derivatives. Then, by using the connections between Frechet derivatives and Mordukhovich derivatives (or coderivatives) of single-valued mappings in Banach spaces, we derive the useful rules for calculating the Mordukhovich derivatives of single-valued mappings in Euclidean spaces. For practicing these rules, we find the precise solutions of the Frechet derivatives and Mordukhovich derivatives for some single-valued mappings in Euclidean spaces (in R^2, it can be extended to R^n). By using these solutions, we will find the covering constants for the considered mappings. As applications of the results about the covering constants and by applying the Arutyunov Mordukhovich and Zhukovskiy Parameterized Coincidence Point Theorem, we solve some parameterized equations.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 15:49:03 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 17:16:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Li",
"Jinlu",
""
]
] |
2508.11118
|
Jinlu Li
|
Jinlu Li
|
Frechet and Mordukhovich Derivative (Coderivative) and Covering Constant for Single-Valued Mapping in Euclidean Space with Application (II)
|
This is a new research article. arXiv admin note: substantial text overlap with arXiv:2508.10766
| null | null | null |
math.FA
|
http://creativecommons.org/licenses/by/4.0/
|
We continue the study in part I for calculating the Frechet derivatives and Mordukhovich derivatives (coderivatives) and covering constants for single-valued mappings in Euclidean spaces (It is part I). In this paper, we particularly consider a norm-reserved mapping f: R^2 to R^2 that is defined by (1.1) in Section 1. We will find the precise solutions of Frechet derivative and Mordukhovich derivative at every point in R^2. By using these solutions, we will find the covering constant for this mapping f is exact 1 at every point in R^2 except the origin. Then we extend this mapping to R^4. Finally, by using the covering constant for f and by applying the Arutyunov Mordukhovich and Zhukovskiy Parameterized Coincidence Point Theorem, we will solve some parameterized equations.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 23:51:20 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 17:58:24 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Li",
"Jinlu",
""
]
] |
2508.11120
|
Lorenzo Jaime Flores
|
Lorenzo Jaime Yu Flores, Junyi Shen, Goodman Gu
|
Towards Reliable Multi-Agent Systems for Marketing Applications via Reflection, Memory, and Planning
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in large language models (LLMs) enabled the development of AI agents that can plan and interact with tools to complete complex tasks. However, literature on their reliability in real-world applications remains limited. In this paper, we introduce a multi-agent framework for a marketing task: audience curation. To solve this, we introduce a framework called RAMP that iteratively plans, calls tools, verifies the output, and generates suggestions to improve the quality of the audience generated. Additionally, we equip the model with a long-term memory store, which is a knowledge base of client-specific facts and past queries. Overall, we demonstrate the use of LLM planning and memory, which increases accuracy by 28 percentage points on a set of 88 evaluation queries. Moreover, we show the impact of iterative verification and reflection on more ambiguous queries, showing progressively better recall (roughly +20 percentage points) with more verify/reflect iterations on a smaller challenge set, and higher user satisfaction. Our results provide practical insights for deploying reliable LLM-based systems in dynamic, industry-facing environments.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 23:52:39 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 22:18:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Flores",
"Lorenzo Jaime Yu",
""
],
[
"Shen",
"Junyi",
""
],
[
"Gu",
"Goodman",
""
]
] |
2508.11552
|
Alice K. Harding
|
Alice K. Harding, Zorawar Wadiasingh and Matthew G. Baring
|
Pair Cascades in Magnetar Magnetospheres
|
29 pages, 14 figures, accepted for publication in Astrophysical Journal
| null | null | null |
astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
Resonant inverse Compton scattering (RICS) of soft thermal photons by relativistic particles on closed magnetic field loops has been proposed to explain the hard emission observed up to, and beyond, 200 keV from magnetars. If particles injected at the base of the loops have Lorentz factors >= 10^2, the RICS spectra will be attenuated by both one-photon pair production and photon splitting in the ultra-strong magnetar fields, producing additional spectral components from pair synchrotron radiation and split photons that produce further generations of pairs and split photons. We investigate such cascades initiated by the primary injected electrons through a Monte Carlo simulation and study the cascade spectra and pair distributions. For most observer angles, the pair synchrotron and split photon spectra dominate the RICS primary spectra and produce complex polarization signals. In particular, the synchrotron spectra are highly polarized with degree 40% - 80%, are softer than the RICS spectra and may account for the high polarization of some magnetar spectra observed by IXPE above 3 keV.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 15:54:06 GMT"
},
{
"version": "v2",
"created": "Mon, 18 Aug 2025 23:23:42 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Harding",
"Alice K.",
""
],
[
"Wadiasingh",
"Zorawar",
""
],
[
"Baring",
"Matthew G.",
""
]
] |
2508.11680
|
Aditya Akella
|
Aditya Akella, Jonathan Farah
|
Comparative Analysis of Time Series Foundation Models for Demographic Forecasting: Enhancing Predictive Accuracy in US Population Dynamics
|
6 pages, 4 figures, 3 tables
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Demographic shifts, influenced by globalization, economic conditions, geopolitical events, and environmental factors, pose significant challenges for policymakers and researchers. Accurate demographic forecasting is essential for informed decision-making in areas such as urban planning, healthcare, and economic policy. This study explores the application of time series foundation models to predict demographic changes in the United States using datasets from the U.S. Census Bureau and Federal Reserve Economic Data (FRED). We evaluate the performance of the Time Series Foundation Model (TimesFM) against traditional baselines including Long Short-Term Memory (LSTM) networks, Autoregressive Integrated Moving Average (ARIMA), and Linear Regression. Our experiments across six demographically diverse states demonstrate that TimesFM achieves the lowest Mean Squared Error (MSE) in 86.67% of test cases, with particularly strong performance on minority populations with sparse historical data. These findings highlight the potential of pre-trained foundation models to enhance demographic analysis and inform proactive policy interventions without requiring extensive task-specific fine-tuning.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 07:55:09 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 07:06:02 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Akella",
"Aditya",
""
],
[
"Farah",
"Jonathan",
""
]
] |
2508.11682
|
Md Basit Azam
|
Md Basit Azam and Sarangthem Ibotombi Singh
|
Age-Normalized HRV Features for Non-Invasive Glucose Prediction: A Pilot Sleep-Aware Machine Learning Study
| null | null | null | null |
eess.SP cs.AI cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Non-invasive glucose monitoring remains a critical challenge in the management of diabetes. HRV during sleep shows promise for glucose prediction however, age-related autonomic changes significantly confound traditional HRV analyses. We analyzed 43 subjects with multi-modal data including sleep-stage specific ECG, HRV features, and clinical measurements. A novel age-normalization technique was applied to the HRV features by, dividing the raw values by age-scaled factors. BayesianRidge regression with 5-fold cross-validation was employed for log-glucose prediction. Age-normalized HRV features achieved R2 = 0.161 (MAE = 0.182) for log-glucose prediction, representing a 25.6% improvement over non-normalized features (R2 = 0.132). The top predictive features were hrv rem mean rr age normalized (r = 0.443, p = 0.004), hrv ds mean rr age normalized (r = 0.438, p = 0.005), and diastolic blood pressure (r = 0.437, p = 0.005). Systematic ablation studies confirmed age-normalization as the critical component, with sleep-stage specific features providing additional predictive value. Age-normalized HRV features significantly enhance glucose prediction accuracy compared with traditional approaches. This sleep-aware methodology addresses fundamental limitations in autonomic function assessment and suggests a preliminary feasibility for non-invasive glucose monitoring applications. However, these results require validation in larger cohorts before clinical consideration.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 19:24:31 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 13:01:13 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Azam",
"Md Basit",
""
],
[
"Singh",
"Sarangthem Ibotombi",
""
]
] |
2508.11760
|
Saurabh Saurabh
|
Saurabh, Maciek Wielgus, Arman Tursunov, Andrei P. Lobanov, and Razieh Emami
|
Semi-analytic studies of accretion disk and magnetic field geometry in M87*
|
12 pages and 8 figures. Comments are welcome!
| null | null | null |
astro-ph.HE gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
Context: Magnetic fields play a pivotal role in dynamics of black hole accretion flows and formation of relativistic jets. Observations by the Event Horizon Telescope (EHT) provided unprecedented insights into accretion structures near black holes. Interpreting these observations requires a theoretical framework linking polarized emission to underlying system properties and magnetic field geometries. Aims: We investigate how system properties, particularly magnetic field geometry in the event horizon scale region, influence the structure of the observable synchrotron emission in M87*. Specifically, we aim to quantify the sensitivity of observables used by the EHT to black hole spin, plasma dynamics, accretion disk thickness, and magnetic field geometry. Methods: We adopt a semi-analytic radiatively inefficient accretion flow model in Kerr spacetime. We vary magnetic field geometry, black hole spin, accretion disk dynamics, and geometric thickness of the disk. We perform general relativistic ray tracing with a full polarized radiative transfer to obtain synthetic images of M87*. We extract EHT observables, such as disk diameter, asymmetry, and polarimetric metrics from synthetic models. We also consider a number of general relativistic magnetohydrodynamics simulations and compare them with the semi-analytical models. Results: We see limited impact of the disk thickness on observables. On the other hand, toroidal field dominated and poloidal field dominated magnetic configurations can be distinguished reliably. The flow dynamics, in particular presence of radial inflow, also significantly impacts the EHT observables. Conclusions: The M87* system is most consistent with a poloidal magnetic field dominated flow with partially radial inflow. While the spin remain elusive, moderate or large positive values are preferred.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 18:11:02 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Saurabh",
"",
""
],
[
"Wielgus",
"Maciek",
""
],
[
"Tursunov",
"Arman",
""
],
[
"Lobanov",
"Andrei P.",
""
],
[
"Emami",
"Razieh",
""
]
] |
2508.11841
|
Bo Chen
|
Bo Chen and Zhi L\"u
|
On the homology description of equivariant unoriented bordism groups
|
26 pages, v2 with minor typos corrected
| null | null | null |
math.AT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We construct a chain complex $\mathfrak{B}$ based on a double complex derived from the universal complex $X(\mathbb{Z}_2^n)$. It is shown that $\mathfrak{B}$ has a nontrivial homology only in degree $n-2$, which is isomorphic to the equivariant unoriented bordism group $\mathcal{Z}_{n+1}(\mathbb{Z}_2^n)$ of all $(n+1)$-dimensional smooth closed $\mathbb{Z}_2^n$-manifolds with isolated fixed points. By analyzing the spectral sequence of $\mathfrak{B}$, we derive a dimension formula for $\mathcal{Z}_{n+1}(\mathbb{Z}_2^n)$ as a $\mathbb{Z}_2$-vector space, which agrees with a recent result for $n=3$.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 23:41:19 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 10:50:19 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chen",
"Bo",
""
],
[
"Lü",
"Zhi",
""
]
] |
2508.11845
|
Marius Miron
|
Marius Miron, David Robinson, Milad Alizadeh, Ellen Gilsenan-McMahon, Gagan Narula, Emmanuel Chemla, Maddie Cusimano, Felix Effenberger, Masato Hagiwara, Benjamin Hoffman, Sara Keen, Diane Kim, Jane Lawton, Jen-Yu Liu, Aza Raskin, Olivier Pietquin, Matthieu Geist
|
What Matters for Bioacoustic Encoding
| null | null | null | null |
cs.SD cs.AI cs.IR cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Bioacoustics, the study of sounds produced by living organisms, plays a vital role in conservation, biodiversity monitoring, and behavioral studies. Many tasks in this field, such as species, individual, and behavior classification and detection, are well-suited to machine learning. However, they often suffer from limited annotated data, highlighting the need for a general-purpose bioacoustic encoder capable of extracting useful representations for diverse downstream tasks. Such encoders have been proposed before, but are often limited in scope due to a focus on a narrow range of species (typically birds), and a reliance on a single model architecture or training paradigm. Moreover, they are usually evaluated on a small set of tasks and datasets. In this work, we present a large-scale empirical study that covers aspects of bioacoustics that are relevant to research but have previously been scarcely considered: training data diversity and scale, model architectures and training recipes, and the breadth of evaluation tasks and datasets. We obtain encoders that are state-of-the-art on the existing and proposed benchmarks. We also identify what matters for training these encoders, such that this work can be extended when more data are available or better architectures are proposed. Specifically, across 26 datasets with tasks including species classification, detection, individual ID, and vocal repertoire discovery, we find self-supervised pre-training followed by supervised post-training on a mixed bioacoustics + general-audio corpus yields the strongest in- and out-of-distribution performance. We show the importance of data diversity in both stages. To support ongoing research and application, we will release the model checkpoints.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 23:52:34 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 12:07:19 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Miron",
"Marius",
""
],
[
"Robinson",
"David",
""
],
[
"Alizadeh",
"Milad",
""
],
[
"Gilsenan-McMahon",
"Ellen",
""
],
[
"Narula",
"Gagan",
""
],
[
"Chemla",
"Emmanuel",
""
],
[
"Cusimano",
"Maddie",
""
],
[
"Effenberger",
"Felix",
""
],
[
"Hagiwara",
"Masato",
""
],
[
"Hoffman",
"Benjamin",
""
],
[
"Keen",
"Sara",
""
],
[
"Kim",
"Diane",
""
],
[
"Lawton",
"Jane",
""
],
[
"Liu",
"Jen-Yu",
""
],
[
"Raskin",
"Aza",
""
],
[
"Pietquin",
"Olivier",
""
],
[
"Geist",
"Matthieu",
""
]
] |
2508.11855
|
Shinji Ohno
|
Shinji Ohno, Takashi Sakai
|
Polars and antipodal sets of generalized $s$-manifolds
| null | null | null | null |
math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we introduce the notion of generalized $s$-manifolds, which is a generalization of symmetric spaces. Then we give a method to construct generalized $s$-manifolds and present some typical examples. We study polars and antipodal sets of generalized $s$-manifolds aiming to extend the results on compact symmetric spaces due to Chen and Nagano.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 00:43:15 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 10:59:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ohno",
"Shinji",
""
],
[
"Sakai",
"Takashi",
""
]
] |
2508.11922
|
Aditi Jahagirdar
|
Aditi Jahagirdar, Sameer Joshi
|
Assessment of Using Synthetic Data in Brain Tumor Segmentation
|
Updates include improved references, clearer table column title, and minor language corrections
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Manual brain tumor segmentation from MRI scans is challenging due to tumor heterogeneity, scarcity of annotated data, and class imbalance in medical imaging datasets. Synthetic data generated by generative models has the potential to mitigate these issues by improving dataset diversity. This study investigates, as a proof of concept, the impact of incorporating synthetic MRI data, generated using a pre-trained GAN model, into training a U-Net segmentation network. Experiments were conducted using real data from the BraTS 2020 dataset, synthetic data generated with the medigan library, and hybrid datasets combining real and synthetic samples in varying proportions. While overall quantitative performance (Dice coefficient, IoU, precision, recall, accuracy) was comparable between real-only and hybrid-trained models, qualitative inspection suggested that hybrid datasets, particularly with 40% real and 60% synthetic data, improved whole tumor boundary delineation. However, region-wise accuracy for the tumor core and the enhancing tumor remained lower, indicating a persistent class imbalance. The findings support the feasibility of synthetic data as an augmentation strategy for brain tumor segmentation, while highlighting the need for larger-scale experiments, volumetric data consistency, and mitigating class imbalance in future work.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 05:56:38 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 16:35:17 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Jahagirdar",
"Aditi",
""
],
[
"Joshi",
"Sameer",
""
]
] |
2508.11930
|
Shanshan Han
|
Shanshan Han, Rongsheng Li, Xingxing Jiang, Ming Lu, Shuxiang Xu, Jiang Zeng, Dong Wu, Nanlin Wang
|
Mn4Al11: A Half-Semimetal Candidate with Anomalous Electronic Behaviors
| null |
Chin. Phys. Lett. 2025
| null |
2025-0843
|
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
Half-semimetals, characterized by their spin-polarized electronic states, hold significant promise for spintronic applications but remain scarce due to stringent electronic and magnetic criteria. Through a combination of transport measurements and optical spectroscopy, we investigated the intermetallic compound Mn4Al11, which features an exceptionally low carrier concentration and undergoes a magnetic phase transition near 68 K. Transport measurements reveal anomalies that deviate from typical metallic behavior at low temperatures. Optical spectroscopy indicates a small, nearly frequency-independent optical conductivity in the far-infrared region, with spectral weight decreasing as the temperature drops from 300 K to 50 K. These behaviors suggest a temperaturedependent carrier density and significant scattering of charge carriers. Combining experimental findings with calculated electronic band structures, we propose that Mn4Al11 is a novel half-semimetal candidate exhibiting a ferrimagnetic ground state.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 06:20:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Han",
"Shanshan",
""
],
[
"Li",
"Rongsheng",
""
],
[
"Jiang",
"Xingxing",
""
],
[
"Lu",
"Ming",
""
],
[
"Xu",
"Shuxiang",
""
],
[
"Zeng",
"Jiang",
""
],
[
"Wu",
"Dong",
""
],
[
"Wang",
"Nanlin",
""
]
] |
2508.11934
|
Xiao-Xiong Zeng
|
Ke Wang, Xiao-Xiong Zeng
|
Energy extraction from the accelerating Kerr black hole via magnetic reconnection in the plunging region and circular orbit region
|
17 pages, 9 figures, minor revisions
| null | null | null |
gr-qc astro-ph.HE
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Based on the magnetic reconnection mechanism, this study investigates how to extract energy effectively from an accelerating Kerr black hole in the plunging region and circular orbit region. After introducing the properties of accelerating black holes, including the event horizon, ergosphere, circular orbits, and innermost stable circular orbit, we investigate the magnetic reconnection process in the plunging region. Specifically, we analyze variations of the azimuthal angle with respect to the acceleration, examine changes in energy per enthalpy of decelerated plasma, and plot energy extraction efficiency along with permissible energy extraction regions. Results show that in the plunging region, at larger radii of reconnection locations, the accelerating black hole exhibits higher energy extraction efficiency than a Kerr black hole. Away from extremality, the acceleration parameter impedes energy extraction, while near extremality, it enhances extraction. We also study energy extraction in circular orbit region by plotting energy extraction efficiency within permissible regions. We find that the permissible energy extraction area is reduced and the efficiency exceeds that of Kerr black holes due to the existence of acceleration parameter. Larger acceleration parameters yield more effective energy extraction regardless of extremality, which is different from that in the plunging region. Additionally, energy extraction efficiency in the plunging region surpasses that in the circular orbit region, aligning with prior conclusions.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 06:29:27 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 10:02:13 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Ke",
""
],
[
"Zeng",
"Xiao-Xiong",
""
]
] |
2508.11948
|
Dae Yeon Kim
|
Dae Yeon Kim, Sachit G. Nagella, Saksham Malik, Nayeon Park, Jaewook Nam, Eric S.G. Shaqfeh, and Sho C. Takatori
|
Colloidal hydrodynamic interactions in viscoelastic fluids
|
22 pages, 7 figures. Supplementary Information and videos available as ancillary files
| null | null | null |
cond-mat.soft
|
http://creativecommons.org/licenses/by/4.0/
|
The motion of suspended colloidal particles generates fluid disturbances in the surrounding medium that create interparticle interactions. While such colloidal hydrodynamic interactions (HIs) have been extensively studied in viscous Newtonian media, comprehensive understanding of HIs in viscoelastic fluids is lacking. We develop a framework to quantify HIs in viscoelastic fluids with high spatiotemporal precision by trapping colloids and inducing translation-rotation hydrodynamic coupling. Using solutions of wormlike micelles (WLMs) as a case study, we discover that HIs are strongly time-dependent and depend on the structural memory generated in the viscoelastic fluid, in contrast to "instantaneous" HIs in viscous Newtonian fluids. We directly measure time-dependent HIs between a stationary probe and a driven particle during transient start-up, developing on the WLM relaxation timescale. Following the sudden cessation of the driven particle, we observe an intriguing flow reversal in the opposing direction, lasting for a time about ten times larger than the WLM relaxation time. We corroborate our observations with analytical microhydrodynamic theory, direct numerical solutions of a continuum model, and particle-based Stokesian dynamics simulations. We find that the structural recovery of the WLMs from a nonlinear strain can generate anisotropic and heterogeneous stresses that produce flow reversals and hydrodynamic attraction among colloids. Measured heterogeneities indicate a breakdown of standard continuum models for constitutive relations when the size of colloids is comparable to the length scales of the polymeric constituents and their entanglement lengths.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 07:22:41 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 03:19:12 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Kim",
"Dae Yeon",
""
],
[
"Nagella",
"Sachit G.",
""
],
[
"Malik",
"Saksham",
""
],
[
"Park",
"Nayeon",
""
],
[
"Nam",
"Jaewook",
""
],
[
"Shaqfeh",
"Eric S. G.",
""
],
[
"Takatori",
"Sho C.",
""
]
] |
2508.11987
|
Jiashuo Liu
|
Zhiyuan Zeng, Jiashuo Liu, Siyuan Chen, Tianci He, Yali Liao, Jinpeng Wang, Zaiyuan Wang, Yang Yang, Lingyue Yin, Mingren Yin, Zhenwei Zhu, Tianle Cai, Zehui Chen, Jiecao Chen, Yantao Du, Xiang Gao, Jiacheng Guo, Liang Hu, Jianpeng Jiao, Xiangsheng Li, Jingkai Liu, Shuang Ni, Zhoufutu Wen, Ge Zhang, Kaiyuan Zhang, Xin Zhou, Jose Blanchet, Xipeng Qiu, Mengdi Wang, Wenhao Huang
|
FutureX: An Advanced Live Benchmark for LLM Agents in Future Prediction
|
Technical report, 51 pages. Update the results
| null | null | null |
cs.AI cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Future prediction is a complex task for LLM agents, requiring a high level of analytical thinking, information gathering, contextual understanding, and decision-making under uncertainty. Agents must not only gather and interpret vast amounts of dynamic information but also integrate diverse data sources, weigh uncertainties, and adapt predictions based on emerging trends, just as human experts do in fields like politics, economics, and finance. Despite its importance, no large-scale benchmark exists for evaluating agents on future prediction, largely due to challenges in handling real-time updates and retrieving timely, accurate answers. To address this, we introduce $\textbf{FutureX}$, a dynamic and live evaluation benchmark specifically designed for LLM agents performing future prediction tasks. FutureX is the largest and most diverse live benchmark for future prediction, supporting real-time daily updates and eliminating data contamination through an automated pipeline for question gathering and answer collection. We evaluate 25 LLM/agent models, including those with reasoning, search capabilities, and integration of external tools such as the open-source Deep Research Agent and closed-source Deep Research models. This comprehensive evaluation assesses agents' adaptive reasoning and performance in dynamic environments. Additionally, we provide in-depth analyses of agents' failure modes and performance pitfalls in future-oriented tasks, including the vulnerability to fake web pages and the temporal validity. Our goal is to establish a dynamic, contamination-free evaluation standard that drives the development of LLM agents capable of performing at the level of professional human analysts in complex reasoning and predictive thinking.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 08:54:08 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 01:46:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zeng",
"Zhiyuan",
""
],
[
"Liu",
"Jiashuo",
""
],
[
"Chen",
"Siyuan",
""
],
[
"He",
"Tianci",
""
],
[
"Liao",
"Yali",
""
],
[
"Wang",
"Jinpeng",
""
],
[
"Wang",
"Zaiyuan",
""
],
[
"Yang",
"Yang",
""
],
[
"Yin",
"Lingyue",
""
],
[
"Yin",
"Mingren",
""
],
[
"Zhu",
"Zhenwei",
""
],
[
"Cai",
"Tianle",
""
],
[
"Chen",
"Zehui",
""
],
[
"Chen",
"Jiecao",
""
],
[
"Du",
"Yantao",
""
],
[
"Gao",
"Xiang",
""
],
[
"Guo",
"Jiacheng",
""
],
[
"Hu",
"Liang",
""
],
[
"Jiao",
"Jianpeng",
""
],
[
"Li",
"Xiangsheng",
""
],
[
"Liu",
"Jingkai",
""
],
[
"Ni",
"Shuang",
""
],
[
"Wen",
"Zhoufutu",
""
],
[
"Zhang",
"Ge",
""
],
[
"Zhang",
"Kaiyuan",
""
],
[
"Zhou",
"Xin",
""
],
[
"Blanchet",
"Jose",
""
],
[
"Qiu",
"Xipeng",
""
],
[
"Wang",
"Mengdi",
""
],
[
"Huang",
"Wenhao",
""
]
] |
2508.12063
|
Denisa Martonov\'a
|
Denisa Martonov\'a, Alain Goriely, Ellen Kuhl
|
Generalized invariants meet constitutive neural networks: A novel framework for hyperelastic materials
| null | null | null | null |
cond-mat.soft cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The major challenge in determining a hyperelastic model for a given material is the choice of invariants and the selection how the strain energy function depends functionally on these invariants. Here we introduce a new data-driven framework that simultaneously discovers appropriate invariants and constitutive models for isotropic incompressible hyperelastic materials. Our approach identifies both the most suitable invariants in a class of generalized invariants and the corresponding strain energy function directly from experimental observations. Unlike previous methods that rely on fixed invariant choices or sequential fitting procedures, our method integrates the discovery process into a single neural network architecture. By looking at a continuous family of possible invariants, the model can flexibly adapt to different material behaviors. We demonstrate the effectiveness of this approach using popular benchmark datasets for rubber and brain tissue. For rubber, the method recovers a stretch-dominated formulation consistent with classical models. For brain tissue, it identifies a formulation sensitive to small stretches, capturing the nonlinear shear response characteristic of soft biological matter. Compared to traditional and neural-network-based models, our framework provides improved predictive accuracy and interpretability across a wide range of deformation states. This unified strategy offers a robust tool for automated and physically meaningful model discovery in hyperelasticity.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 14:32:45 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 16:52:49 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Martonová",
"Denisa",
""
],
[
"Goriely",
"Alain",
""
],
[
"Kuhl",
"Ellen",
""
]
] |
2508.12088
|
Carlos Colosqui Prof.
|
Carlos E. Colosqui
|
Dynamic-Kinetic Duality of Particulate and Multiphase Systems
|
21 pages, 5 figures
| null | null | null |
cond-mat.soft
|
http://creativecommons.org/licenses/by/4.0/
|
The evolution of particulate and multiphase systems can transition from dynamic regimes, governed by classical transport equations with well-defined damping coefficients, to anomalously slow relaxation described by rate equations when the system is critically close to equilibrium. This regime crossover has been both theoretically predicted and experimentally observed in diverse multiphase systems relevant to numerous technological applications, including nanoparticle adhesion at interfaces and liquid imbibition under microscale confinement, and it is attributed to the presence of small nanoscale features of physical or chemical nature at liquid-solid interfaces. This article presents a theoretical framework to more accurately predict and control, advancing or delaying, the dynamic-to-kinetic regime crossover, and highlights strategies for harnessing this phenomenon to enhance or suppress different transport processes in confined multiphase systems.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 15:52:33 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 11:51:56 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Colosqui",
"Carlos E.",
""
]
] |
2508.12219
|
Xiaobo Cai
|
Kaiyuan Wang, Jixing Liu, Xiaobo Cai
|
C2PSA-Enhanced YOLOv11 Architecture: A Novel Approach for Small Target Detection in Cotton Disease Diagnosis
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
This study presents a deep learning-based optimization of YOLOv11 for cotton disease detection, developing an intelligent monitoring system. Three key challenges are addressed: (1) low precision in early spot detection (35% leakage rate for sub-5mm2 spots), (2) performance degradation in field conditions (25% accuracy drop), and (3) high error rates (34.7%) in multi-disease scenarios. The proposed solutions include: C2PSA module for enhanced small-target feature extraction; Dynamic category weighting to handle sample imbalance; Improved data augmentation via Mosaic-MixUp scaling. Experimental results on a 4,078-image dataset show: mAP50: 0.820 (+8.0% improvement); mAP50-95: 0.705 (+10.5% improvement); Inference speed: 158 FPS. The mobile-deployed system enables real-time disease monitoring and precision treatment in agricultural applications.
|
[
{
"version": "v1",
"created": "Sun, 17 Aug 2025 03:28:02 GMT"
},
{
"version": "v2",
"created": "Tue, 19 Aug 2025 00:54:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Kaiyuan",
""
],
[
"Liu",
"Jixing",
""
],
[
"Cai",
"Xiaobo",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.