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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2508.13326
|
Dylan Cope
|
Dylan Cope, Peter McBurney
|
Decoding Communications with Partial Information
|
Proceedings of ALIFE 2025
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Machine language acquisition is often presented as a problem of imitation learning: there exists a community of language users from which a learner observes speech acts and attempts to decode the mappings between utterances and situations. However, an interesting consideration that is typically unaddressed is partial observability, i.e. the learner is assumed to see all relevant information. This paper explores relaxing this assumption, thereby posing a more challenging setting where such information needs to be inferred from knowledge of the environment, the actions taken, and messages sent. We see several motivating examples of this problem, demonstrate how they can be solved in a toy setting, and formally explore challenges that arise in more general settings. A learning-based algorithm is then presented to perform the decoding of private information to facilitate language acquisition.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:19:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cope",
"Dylan",
""
],
[
"McBurney",
"Peter",
""
]
] |
2508.13327
|
Sarthak Khanna
|
Sarthak Khanna, Armin Berger, David Berghaus, Tobias Deusser, Lorenz Sparrenberg, Rafet Sifa
|
Towards Unified Multimodal Financial Forecasting: Integrating Sentiment Embeddings and Market Indicators via Cross-Modal Attention
|
Accepted in IEEE-DSAA 2025
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose STONK (Stock Optimization using News Knowledge), a multimodal framework integrating numerical market indicators with sentiment-enriched news embeddings to improve daily stock-movement prediction. By combining numerical & textual embeddings via feature concatenation and cross-modal attention, our unified pipeline addresses limitations of isolated analyses. Backtesting shows STONK outperforms numeric-only baselines. A comprehensive evaluation of fusion strategies and model configurations offers evidence-based guidance for scalable multimodal financial forecasting. Source code is available on GitHub
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:22:39 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Khanna",
"Sarthak",
""
],
[
"Berger",
"Armin",
""
],
[
"Berghaus",
"David",
""
],
[
"Deusser",
"Tobias",
""
],
[
"Sparrenberg",
"Lorenz",
""
],
[
"Sifa",
"Rafet",
""
]
] |
2508.13328
|
Zeinab Davarani
|
Amirali Arbab, Zeinab Davarani, Mehran Safayani
|
A Dual-Attention Graph Network for fMRI Data Classification
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Understanding the complex neural activity dynamics is crucial for the development of the field of neuroscience. Although current functional MRI classification approaches tend to be based on static functional connectivity or cannot capture spatio-temporal relationships comprehensively, we present a new framework that leverages dynamic graph creation and spatiotemporal attention mechanisms for Autism Spectrum Disorder(ASD) diagnosis. The approach used in this research dynamically infers functional brain connectivity in each time interval using transformer-based attention mechanisms, enabling the model to selectively focus on crucial brain regions and time segments. By constructing time-varying graphs that are then processed with Graph Convolutional Networks (GCNs) and transformers, our method successfully captures both localized interactions and global temporal dependencies. Evaluated on the subset of ABIDE dataset, our model achieves 63.2 accuracy and 60.0 AUC, outperforming static graph-based approaches (e.g., GCN:51.8). This validates the efficacy of joint modeling of dynamic connectivity and spatio-temporal context for fMRI classification. The core novelty arises from (1) attention-driven dynamic graph creation that learns temporal brain region interactions and (2) hierarchical spatio-temporal feature fusion through GCNtransformer fusion.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:23:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Arbab",
"Amirali",
""
],
[
"Davarani",
"Zeinab",
""
],
[
"Safayani",
"Mehran",
""
]
] |
2508.13329
|
Abel-John Buchner
|
Guillaume Ricard, Gunnar Jacobi, Daniele Fiscaletti, Abel-John Buchner
|
Dynamic stall of a hydrofoil with tubercles in surface gravity waves
| null | null | null | null |
physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
The interaction of an object with an unsteady flow is non-trivial and is still far from being fully understood. When an airfoil or hydrofoil, for example, undergoes time-dependent motion, nonlinear flow phenomena such as dynamic stall can emerge. The present work experimentally investigates the interaction between a hydrofoil and surface gravity waves. The waves impose periodic fluctuations of the velocity magnitude and orientation, causing a steadily translating hydrofoil to be susceptible to dynamic stall at large wave forcing amplitudes. Simultaneous measurement of both the forces acting on the hydrofoil and the flow around it by means of particle image velocimetry (PIV) are performed, to properly characterise the hydrofoil-wave interaction. In an attempt at alleviating the impact of the flow unsteadiness via passive flow control, a bio-inspired tubercle geometry is applied along the hydrofoil leading edge. This geometry is known to delay stall in steady cases but has scarcely been studied in unsteady flow conditions. The vortex structures associated with dynamic stall are identified, and their trajectories, dimension, and strength characterised. This analysis is performed for both straight- and tubercled-leading-edge geometries, with tubercles found to qualitatively modify the flow behaviour during dynamic stall. Contrary to previous studies, direct measurements of lift do not evidence any strong modification by tubercles. Drag-driven horizontal force fluctuations, however, which have not previously been measured in this context, are found to be strongly attenuated. This decrease is quantified, and a physical model based on the flow observations is finally proposed.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:25:43 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ricard",
"Guillaume",
""
],
[
"Jacobi",
"Gunnar",
""
],
[
"Fiscaletti",
"Daniele",
""
],
[
"Buchner",
"Abel-John",
""
]
] |
2508.13330
|
Mojtaba Shahbazi
|
Mojtaba Shahbazi, Mehdi Sadeghi
|
Callan-Symanzik-like equation in information theory
|
12 pages, no figure
| null | null | null |
quant-ph hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we consider the jumps in the complexity growth rate (CGR) through the complexity=anything proposal. It is revealed that bulk fields are in charge of the height and the location of these jumps. To put it accurately, their counterparts are divergence of energy momentum tensor, shear viscosity and the Weyl anomaly of the boundary theory. The behavior of CGR around these jumps manifests a critical behavior and critical exponents. Moreover, it is figured out that CGR satisfies a Callan-Symanzik-like equation around these jumps. This equation suggests that the speed of information processing could be changed by the scale of energy. Put it another way, by change of the scale of energy one could turn the classical computers into faster processor computers even faster than our current quantum computers.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:25:43 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Shahbazi",
"Mojtaba",
""
],
[
"Sadeghi",
"Mehdi",
""
]
] |
2508.13331
|
Sam Nelson
|
Sam Nelson and Natsumi Oyamaguchi
|
Psyquandle Brackets
|
14 pages
| null | null | null |
math.GT math.QA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We extend the notion of biquandle brackets to the case of psyquandles, defining quantum enhancements of the psyquandle counting invariant for singular knots and pseudoknots. We provide examples to illustrate the computation of these invariants, establishing that the enhancement is proper. We compute a few toy examples, noting that the true power of this infinite family of invariants lies in more computationally expensive larger-cardinality psyquandles and infinite coefficient rings.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:31:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Nelson",
"Sam",
""
],
[
"Oyamaguchi",
"Natsumi",
""
]
] |
2508.13332
|
Sergio Giardino
|
Sergio Giardino
|
Expectation value dynamics within real Hilbert space quantum mechanics
|
Accpet by Int. J. Theor. Phys
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Dynamic equations concerning physical expectation values have been examined in terms of the real Hilbert space approach to quantum mechanics. The considered cases involve complex wave functions, as well as quaternionic wave functions. The consistency of the formalism has been verified in terms of the continuity equation, the classical limit, and generalizations of the quantum Lorentz force, and the Virial theorem. Besides testing the consistency of the real Hilbert space approach, generalized position and angular momentum operators have been introduced, and inspire exciting directions for further research.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:40:53 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Giardino",
"Sergio",
""
]
] |
2508.13333
|
Chentong Chen
|
Chentong Chen, Mengyuan Zhong, Jianyong Sun, Ye Fan, Jialong Shi
|
HiFo-Prompt: Prompting with Hindsight and Foresight for LLM-based Automatic Heuristic Design
|
9 pages, 6 figures
| null | null | null |
cs.AI cs.NE math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
LLM-based Automatic Heuristic Design (AHD) within Evolutionary Computation (EC) frameworks has shown promising results. However, its effectiveness is hindered by the use of static operators and the lack of knowledge accumulation mechanisms. We introduce HiFo-Prompt, a framework that guides LLMs with two synergistic prompting strategies: Foresight and Hindsight. Foresight-based prompts adaptively steer the search based on population dynamics, managing the exploration-exploitation trade-off. In addition, hindsight-based prompts mimic human expertise by distilling successful heuristics from past generations into fundamental, reusable design principles. This dual mechanism transforms transient discoveries into a persistent knowledge base, enabling the LLM to learn from its own experience. Empirical results demonstrate that HiFo-Prompt significantly outperforms state-of-the-art LLM-based AHD methods, generating higher-quality heuristics while achieving substantially faster convergence and superior query efficiency.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:42:55 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chen",
"Chentong",
""
],
[
"Zhong",
"Mengyuan",
""
],
[
"Sun",
"Jianyong",
""
],
[
"Fan",
"Ye",
""
],
[
"Shi",
"Jialong",
""
]
] |
2508.13334
|
Adrian Ducourtial
|
Adrian Ducourtial
|
Higher arithmetic on the ordinals
|
9 pages
| null | null | null |
math.LO
|
http://creativecommons.org/licenses/by/4.0/
|
We motivate and study an infinite sequence of binary operations on the ordinal numbers, extending the standard arithmetic on the ordinals to higher degrees of iteration. Connections to the hyperoperations on the natural numbers are discussed.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:44:28 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ducourtial",
"Adrian",
""
]
] |
2508.13335
|
Suresh Tiwari dr
|
S C Tiwari
|
Harmonic Spinors and $Z_2$ Vortex
|
9 pages
| null | null | null |
physics.gen-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Hodge theorem and harmonic spinors are studied in a physics-oriented approach in the present paper. New mathematical results on the harmonic spinors are as follows. Harmonic spinors defined by partial differential operators could be of two types: trivial without topological defects, and having nontrivial topological structures, for example, phase singularities or phase vortices. There could exist a nontrivial harmonic vector field associated with nontrivial harmonic spinor, for example, ${\bf v}_{vortex}$ associated with Weyl 2-spinor. The $Z_2$-vortex is re-visited in the perspective of harmonic spinors leading to a remarkable result that the gauge potential is exactly the same as the nontrivial harmonic vector field associated with the 2-spinor. It is proposed that a discrete symmetry group $SL(2, Z)$ has a role in connection with the continuous group $SU(2)$ similar to the discrete group $Z_2$ in $U(1)$.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:45:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Tiwari",
"S C",
""
]
] |
2508.13336
|
Javad Asadzade
|
Javad A. Asadzade and Nazim I. Mahmudov
|
Necessary and sufficient conditions for relative controllability of discrete linear delay systems
| null | null | null | null |
math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper, we investigate delayed linear difference systems and establish several fundamental results. We first provide a Kalman-type rank condition tailored for delayed linear difference systems. Furthermore, we construct the discrete Gramian matrix and prove its non-singularity, which is essential for analyzing system properties. Additionally, we obtain necessary and sufficient conditions ensuring the relative controllability of the system. Finally, we formulate the control function for effective system control and validate our theoretical findings with numerical examples. These examples illustrate the practical behavior of discrete linear delay systems.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:48:39 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Asadzade",
"Javad A.",
""
],
[
"Mahmudov",
"Nazim I.",
""
]
] |
2508.13337
|
Yueming Yuan
|
Yueming Yuan, Ahan Gupta, Jianping Li, Sajal Dash, Feiyi Wang, Minjia Zhang
|
X-MoE: Enabling Scalable Training for Emerging Mixture-of-Experts Architectures on HPC Platforms
|
17 pages, 20 figures. To be published in SC 2025
| null | null | null |
cs.LG cs.CL cs.DC
|
http://creativecommons.org/licenses/by/4.0/
|
Emerging expert-specialized Mixture-of-Experts (MoE) architectures, such as DeepSeek-MoE, deliver strong model quality through fine-grained expert segmentation and large top-k routing. However, their scalability is limited by substantial activation memory overhead and costly all-to-all communication. Furthermore, current MoE training systems - primarily optimized for NVIDIA GPUs - perform suboptimally on non-NVIDIA platforms, leaving significant computational potential untapped. In this work, we present X-MoE, a novel MoE training system designed to deliver scalable training performance for next-generation MoE architectures. X-MoE achieves this via several novel techniques, including efficient padding-free MoE training with cross-platform kernels, redundancy-bypassing dispatch, and hybrid parallelism with sequence-sharded MoE blocks. Our evaluation on the Frontier supercomputer, powered by AMD MI250X GPUs, shows that X-MoE scales DeepSeek-style MoEs up to 545 billion parameters across 1024 GPUs - 10x larger than the largest trainable model with existing methods under the same hardware budget, while maintaining high training throughput. The source code of X-MoE is available at https://github.com/Supercomputing-System-AI-Lab/X-MoE.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:49:28 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Yuan",
"Yueming",
""
],
[
"Gupta",
"Ahan",
""
],
[
"Li",
"Jianping",
""
],
[
"Dash",
"Sajal",
""
],
[
"Wang",
"Feiyi",
""
],
[
"Zhang",
"Minjia",
""
]
] |
2508.13338
|
Manuel Alejandro Mart\'inez Flores
|
Duv\'an Cardona, Manuel Mart\'inez
|
Estimates for pseudo-differential operators on the torus revisited. III
|
19 pages
| null | null | null |
math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
This paper finishes the goal of the authors started in two previous manuscripts dedicated to revisiting the continuity properties of toroidal pseudo-differential operators with symbols in the H\"ormander classes. Here we prove pointwise estimates in terms of the Fefferman-Stein sharp maximal function and of the Hardy-Littlewood maximal function. Combining these estimates with the properties of Muckenhoupt's weight class $A_p$ we obtain boundedness theorems for pseudo-differential operators between weighted Lebesgue spaces on the torus $L^p(w)$. These results are given in the context of the global symbolic analysis defined on $\mathbb{T}^n\times \mathbb{Z}^n$ as developed by Ruzhansky and Turunen by using discrete Fourier analysis, and extend those of Park and Tomita available in the Euclidean case. Moreover, we include continuity results on Sobolev spaces $W^s_p$ and on Besov spaces $B^s_{p,q}$ on the torus. Our techniques are taken from Park and Tomita \cite{park-tomita} and we consider its toroidal extension here for the completeness of the boundedness of toroidal pseudo-differential operators with respect to the current literature.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:53:40 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cardona",
"Duván",
""
],
[
"Martínez",
"Manuel",
""
]
] |
2508.13339
|
Eugene Hamzezadeh
|
Eugene T. Hamzezadeh and Andrew J. Petruska
|
Observed Control -- Linearly Scalable Nonlinear Model Predictive Control with Adaptive Horizons
|
16 pages, 8 figures. Submitted to IEEE Transactions on Automatic Control 8/17/2025
| null | null | null |
math.OC cs.RO cs.SY eess.SY
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This work highlights the duality between state estimation methods and model predictive control. A predictive controller, observed control, is presented that uses this duality to efficiently compute control actions with linear time-horizon length scalability. The proposed algorithms provide exceptional computational efficiency, adaptive time horizon lengths, and early optimization termination criteria. The use of Kalman smoothers as the backend optimization framework provides for a straightforward implementation supported by strong theoretical guarantees. Additionally, a formulation is presented that separates linear model predictive control into purely reactive and anticipatory components, enabling any-time any-horizon observed control while ensuring controller stability for short time horizons. Finally, numerical case studies confirm that nonlinear filter extensions, i.e., the extended Kalman filter and unscented Kalman filter, effectively extend observed control to nonlinear systems and objectives.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:53:44 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hamzezadeh",
"Eugene T.",
""
],
[
"Petruska",
"Andrew J.",
""
]
] |
2508.13340
|
Sedigheh Dargahi
|
Sedigheh Dargahi, Sylvain Bouix, Christian Desrosier
|
Susceptibility Distortion Correction of Diffusion MRI with a single Phase-Encoding Direction
| null | null | null | null |
eess.IV cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Diffusion MRI (dMRI) is a valuable tool to map brain microstructure and connectivity by analyzing water molecule diffusion in tissue. However, acquiring dMRI data requires to capture multiple 3D brain volumes in a short time, often leading to trade-offs in image quality. One challenging artifact is susceptibility-induced distortion, which introduces significant geometric and intensity deformations. Traditional correction methods, such as topup, rely on having access to blip-up and blip-down image pairs, limiting their applicability to retrospective data acquired with a single phase encoding direction. In this work, we propose a deep learning-based approach to correct susceptibility distortions using only a single acquisition (either blip-up or blip-down), eliminating the need for paired acquisitions. Experimental results show that our method achieves performance comparable to topup, demonstrating its potential as an efficient and practical alternative for susceptibility distortion correction in dMRI.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:56:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dargahi",
"Sedigheh",
""
],
[
"Bouix",
"Sylvain",
""
],
[
"Desrosier",
"Christian",
""
]
] |
2508.13341
|
Mohsen Fathi
|
Mohsen Fathi
|
Shadow of a nonlinear electromagnetic generalized Kerr-Newman-AdS black hole
|
22 pages, 11 figures, 4 tables
| null | null | null |
gr-qc astro-ph.HE hep-th
|
http://creativecommons.org/licenses/by/4.0/
|
In this work, we investigate the shadow properties of the Kerr-Newman-Anti-de Sitter black hole coupled to a nonlinear electrodynamics. We construct the shadow by employing the celestial coordinate approach for observers located at finite distances, accounting for the non-asymptotically flat nature of the spacetime. The size, distortion, area, and oblateness of the shadow are analyzed in terms of the black hole parameters, including the spin, effective charge, and nonlinearity parameter. We further explore observational constraints using the Event Horizon Telescope results for M87* and Sgr~A*. The angular diameters inferred from the observed images allow us to bound the parameter space of the model, showing that both spin and nonlinearity play key roles in shaping the shadow and in restricting the effective charge. Additionally, we examine the energy emission rate derived from the shadow radius and Hawking temperature, highlighting the influence of rotation and nonlinear electrodynamics on black hole evaporation. Our results demonstrate that the black hole model provides a physically consistent and observationally viable extension of the standard Kerr paradigm, with potential implications for near-horizon quantum processes and tests of gravity beyond general relativity.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 19:58:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fathi",
"Mohsen",
""
]
] |
2508.13342
|
Mansur Shakipov
|
Ricardo H. Nochetto, Mansur Shakipov
|
Surface Stokes Without Inf-Sup Condition
|
22 pages, 1 figure
| null | null | null |
math.NA cs.NA math.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For a $d$-dimensional hypersurface of class $C^3$ without boundary, we reformulate the surface Stokes equations as a nonsymmetric indefinite elliptic problem governed by two Laplacians. We then use this elliptic reformulation as a basis for a numerical method based on lifted parametric FEM. Assuming no geometric error for simplicity, we prove its well-posedness, quasi-best approximation in a robust mesh-dependent $H^1$-norm for any polynomial degree, as well as an optimal $L^2$ error estimate for both velocity and pressure. This entails a sufficiently small mesh size that solely depends on the Weingarten map and circumvents the usual discrete inf-sup condition. We present numerical experiments for velocity-pressure pairs with equal and disparate polynomial degrees, demonstrating that the proposed method is both accurate and practical.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:01:15 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Nochetto",
"Ricardo H.",
""
],
[
"Shakipov",
"Mansur",
""
]
] |
2508.13343
|
Christian M\"uller
|
Oleg Karpenkov, Christian M\"uller, Anna Pratoussevitch
|
Stressability of Semi-Discrete Frameworks
|
18 pages
| null | null | null |
math.MG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we study the stressability of semi-discrete frameworks in the plane which are generated by a discrete sequence of smooth curves. We characterize their stressability property by the existence of stresses fulfilling certain difference-differential equation. In particular, we define a semi-discrete height function which we use to generate liftings. Furthermore, we show a semi-discrete analogue of the Maxwell-Cremona lifting property which implies that the stressable semi-discrete frameworks in the plane are precisely the orthogonal projections of semi-discrete conjugate surfaces in 3-space. Finally, we discuss geometric implications for frameworks with vanishing boundary forces and characterize the liftability of frameworks which only consist of two neighboring curves forming one strip.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:02:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Karpenkov",
"Oleg",
""
],
[
"Müller",
"Christian",
""
],
[
"Pratoussevitch",
"Anna",
""
]
] |
2508.13344
|
Dolapo Oyetunbi
|
Dolapo Oyetunbi and Dilian Yang
|
Maximality and symmetry related to the \(2\)-adic ring \(C^*\)-algebra
|
Comments are welcome and appreciated
| null | null | null |
math.OA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The 2-adic ring $C^*$-algebra $\mathcal{Q}_2$ is the universal $C^*$-algebra generated by a unitary and an isometry satisfying certain relations. It contains a canonical copy of the Cuntz algebra $\mathcal{O}_2$. We show that $\mathcal{O}_2$ is a maximal $C^*$-subalgebra of $\mathcal{Q}_2$. Furthermore, we examine the structure of the fixed-point algebra under a periodic \(^*\)-automorphism $\sigma$ of $\mathcal{Q}_2$, which is extended from the flip-flop \(^*\)-automorphism of $\mathcal{O}_2$. We show that the maximality of $\mathcal{O}_2$ in $\mathcal{Q}_2$ extends to the crossed product $\mathcal{O}_2 \rtimes_{\sigma} \mathbb{Z}_2$ in $\mathcal{Q}_2 \rtimes_{\sigma} \mathbb{Z}_2$, and to the fixed-point algebra $\mathcal{O}_2^\sigma$ in $\mathcal{Q}_2^\sigma$. As a consequences of our main results, a few open questions concerning $\mathcal{Q}_2$ are resolved.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:03:09 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Oyetunbi",
"Dolapo",
""
],
[
"Yang",
"Dilian",
""
]
] |
2508.13345
|
Joshua Brakensiek
|
Joshua Brakensiek and Venkatesan Guruswami and Aaron Putterman
|
Tight Bounds for Sparsifying Random CSPs
|
67 pages
| null | null | null |
cs.DS cs.DM math.CO
|
http://creativecommons.org/publicdomain/zero/1.0/
|
The problem of CSP sparsification asks: for a given CSP instance, what is the sparsest possible reweighting such that for every possible assignment to the instance, the number of satisfied constraints is preserved up to a factor of $1 \pm \epsilon$? We initiate the study of the sparsification of random CSPs. In particular, we consider two natural random models: the $r$-partite model and the uniform model. In the $r$-partite model, CSPs are formed by partitioning the variables into $r$ parts, with constraints selected by randomly picking one vertex out of each part. In the uniform model, $r$ distinct vertices are chosen at random from the pool of variables to form each constraint.
In the $r$-partite model, we exhibit a sharp threshold phenomenon. For every predicate $P$, there is an integer $k$ such that a random instance on $n$ vertices and $m$ edges cannot (essentially) be sparsified if $m \le n^k$ and can be sparsified to size $\approx n^k$ if $m \ge n^k$. Here, $k$ corresponds to the largest copy of the AND which can be found within $P$. Furthermore, these sparsifiers are simple, as they can be constructed by i.i.d. sampling of the edges.
In the uniform model, the situation is a bit more complex. For every predicate $P$, there is an integer $k$ such that a random instance on $n$ vertices and $m$ edges cannot (essentially) be sparsified if $m \le n^k$ and can sparsified to size $\approx n^k$ if $m \ge n^{k+1}$. However, for some predicates $P$, if $m \in [n^k, n^{k+1}]$, there may or may not be a nontrivial sparsifier. In fact, we show that there are predicates where the sparsifiability of random instances is non-monotone, i.e., as we add more random constraints, the instances become more sparsifiable. We give a precise (efficiently computable) procedure for determining which situation a specific predicate $P$ falls into.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:04:07 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Brakensiek",
"Joshua",
""
],
[
"Guruswami",
"Venkatesan",
""
],
[
"Putterman",
"Aaron",
""
]
] |
2508.13346
|
Daniel Hsu
|
Daniel Hsu
|
Dimension lower bounds for linear approaches to function approximation
|
First appeared on author's homepage in August 2021 https://www.cs.columbia.edu/~djhsu/papers/dimension-argument.pdf
| null | null | null |
cs.LG math.ST stat.TH
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This short note presents a linear algebraic approach to proving dimension lower bounds for linear methods that solve $L^2$ function approximation problems. The basic argument has appeared in the literature before (e.g., Barron, 1993) for establishing lower bounds on Kolmogorov $n$-widths. The argument is applied to give sample size lower bounds for kernel methods.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:04:46 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hsu",
"Daniel",
""
]
] |
2508.13347
|
Waldo G\'alvez
|
Susanne Albers, Waldo G\'alvez, and \"Omer Behic \"Ozdemir
|
On the 2D Demand Bin Packing Problem: Hardness and Approximation Algorithms
|
An extended abstract of this article has been accepted at The XIII Latin American Algorithms, Graphs, and Optimization Symposium (LAGOS 2025)
| null | null | null |
cs.DS
|
http://creativecommons.org/licenses/by/4.0/
|
We study a two-dimensional generalization of the classical Bin Packing problem, denoted as 2D Demand Bin Packing. In this context, each bin is a horizontal timeline, and rectangular tasks (representing electric appliances or computational requirements) must be allocated into the minimum number of bins so that the sum of the heights of tasks at any point in time is at most a given constant capacity. We prove that simple variants of the problem are NP-hard to approximate within a factor better than $2$, namely when tasks have short height and when they are squares, and provide best-possible approximation algorithms for them; we also present a simple $3$-approximation for the general case. All our algorithms are based on a general framework that computes structured solutions for relatively large tasks, while including relatively small tasks on top via a generalization of the well-known First-Fit algorithm for Bin Packing.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:15:38 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Albers",
"Susanne",
""
],
[
"Gálvez",
"Waldo",
""
],
[
"Özdemir",
"Ömer Behic",
""
]
] |
2508.13348
|
Torsten B\"ohm
|
T. B\"ohm, M. Holschneider, P. Petit, F. Ligni\`eres, F. Paletou, C. P. Folsom and M. Rainer
|
Stable magnetic fields and changing starspots on Vega -- An ultra-deep decadal survey at Pic du Midi and OHP
| null | null | null | null |
astro-ph.SR
|
http://creativecommons.org/licenses/by/4.0/
|
Monitoring magnetic and activity variations in A- and B-type stars with ultra-weak magnetic fields is essential to understand the origin and evolution of these fields in this region of the HR diagram, with Vega standing as the prototype of this category. We collected high-resolution spectroscopic and spectropolarimetric data with SOPHIE (OHP, 2018) and with NARVAL/NEO-NARVAL (TBL, 2018, 2023, and 2024), yielding a total of 13108 individual spectra. Magnetic field maps were reconstructed using the Zeeman-Doppler Imaging method, while brightness maps were derived with a dedicated code developed for this purpose. The average magnetic field confirms a negative radial-field spot at the pole with stable strength, while the maps reveal the long-term stability of an oblique dipole together with smaller magnetic structures consistently detected across observing epochs. In contrast, brightness maps show strong variations in the location of surface spots on timescales of years, possibly shorter, although the spot contrast remains nearly unchanged between 2012, 2018, 2023, and 2024, with a normalized spectral amplitude of 0.0003. No direct correlation between magnetic and brightness features could be established in the simultaneous SOPHIE and NARVAL dataset of 2018. These results suggest that Vega hosts both a persistent fossil magnetic field and a dynamo-generated component, most likely concentrated in equatorial regions.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:18:23 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Böhm",
"T.",
""
],
[
"Holschneider",
"M.",
""
],
[
"Petit",
"P.",
""
],
[
"Lignières",
"F.",
""
],
[
"Paletou",
"F.",
""
],
[
"Folsom",
"C. P.",
""
],
[
"Rainer",
"M.",
""
]
] |
2508.13349
|
Kalyan Dey
|
Dipankar Basak, H. Hushnud, Kalyan Dey
|
Harnessing data-driven methods for precise model independent event shape estimation in relativistic heavy-ion collisions
|
10 pages, 7 figures
| null | null | null |
hep-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This study demonstrates the application of supervised machine learning (ML) techniques to distinguish between isotropic and jet-like event topologies in heavy-ion collisions via the spherocity observable. State-of-the-art ML algorithms, optimized through systematic hyperparameter tuning, are employed to predict both traditional transverse spherocity $S_{0}$ and unweighted transverse spherocity $S_{0}^{p_{\rm T}=1}$ directly from raw event data. Moreover, the results from this study demonstrated that our approach remains largely model-independent, underscoring its potential applicability in future experimental heavy-ion physics analyses.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:19:00 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Basak",
"Dipankar",
""
],
[
"Hushnud",
"H.",
""
],
[
"Dey",
"Kalyan",
""
]
] |
2508.13350
|
Raphael Chinchilla
|
Raphael Chinchilla, Thomas D. Rueter, Timothy R. McDade, Peter R. Fisher, Emmanuel Candes, Trevor Hastie, Stephen Boyd
|
Adaptive Strategies for Pension Fund Management
| null | null | null | null |
math.OC q-fin.CP
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This paper proposes a simulation-based framework for assessing and improving the performance of a pension fund management scheme. This framework is modular and allows the definition of customized performance metrics that are used to assess and iteratively improve asset and liability management policies. We illustrate our framework with a simple implementation that showcases the power of including adaptable features. We show that it is possible to dissipate longevity and volatility risks by permitting adaptability in asset allocation and payout levels. The numerical results show that by including a small amount of flexibility, there can be a substantial reduction in the cost to run the pension plan as well as a substantial decrease in the probability of defaulting.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:19:42 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chinchilla",
"Raphael",
""
],
[
"Rueter",
"Thomas D.",
""
],
[
"McDade",
"Timothy R.",
""
],
[
"Fisher",
"Peter R.",
""
],
[
"Candes",
"Emmanuel",
""
],
[
"Hastie",
"Trevor",
""
],
[
"Boyd",
"Stephen",
""
]
] |
2508.13351
|
Katie Bechtold
|
Katie Bechtold, Torsten B\"oker, David E. Franz, Dennis Garland, Maurice te Plate, Timothy D. Rawle, Christopher Q. Trinh, Rai Wu, Peter Zeidler
|
Evolution of JWST Contingency Payload Operations for Mitigating NIRSpec Micro-shutter Array Electrical Shorts
|
14 pages, 9 figures, 18th International Conference on Space Operations, Montreal, Canada, 26 - 30 May 2025, https://star.spaceops.org/2025/user_manudownload.php?doc=428__c7nsvx95.pdf
| null | null | null |
astro-ph.IM
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The Near-Infrared Spectrograph (NIRSpec) is one of four science instruments on board the James Webb Space Telescope (JWST), which began routine operations in July 2022. As JWST's primary spectroscopic instrument for faint, distant targets, NIRSpec plays a central role in several of the mission's core science goals. Its signature multi-object spectroscopy (MOS) mode enables the simultaneous acquisition of spectra for up to a hundred targets across the field of view, using a micro-shutter array (MSA) comprised of nearly 250,000 individually addressable micro-electromechanical shutters.
The MSA is susceptible to occasional electrical shorts, which produce unwanted infrared glow in MOS exposures, rendering them unusable and wasting valuable observatory time. Mitigation requires promptly identifying the affected shutter(s) and masking the corresponding row(s) or column(s) to prevent future activation. However, masking shutters unnecessarily reduces NIRSpec's multiplexing capacity, so it is important to minimize the extent of masking while reliably suppressing the short's effects.
More than two years of operations have informed refinements in technical procedures and operational decision-making around short mitigation. This experience also supports new efforts to identify and unmask previously affected rows or columns where shorts have since disappeared, enabling the recovery of multiplexing capacity that would otherwise remain lost.
As MSA technology matures toward use in future missions such as the Habitable Worlds Observatory, operational strategies developed for NIRSpec remain relevant beyond JWST. This paper provides an overview of MSA shorts mitigation operations, summarizes shorts observed since the completion of commissioning, and describes the evolving contingency procedures that have enabled continued science productivity from this unique payload.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:19:57 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bechtold",
"Katie",
""
],
[
"Böker",
"Torsten",
""
],
[
"Franz",
"David E.",
""
],
[
"Garland",
"Dennis",
""
],
[
"Plate",
"Maurice te",
""
],
[
"Rawle",
"Timothy D.",
""
],
[
"Trinh",
"Christopher Q.",
""
],
[
"Wu",
"Rai",
""
],
[
"Zeidler",
"Peter",
""
]
] |
2508.13352
|
David Cruz-Uribe OFS
|
David Cruz-Uribe
|
Recent results on matrix weighted norm inequalities
|
This paper is a substantially revised version of a previous paper: see arXiv:2304.03887
| null | null | null |
math.CA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this paper we give an overview of recent work on matrix weights, with particular emphasis on convex body sparse domination for singular integrals, Rubio de Francia extrapolation, and Jones factorization. To provide context and motivation, we survey the comparable results in the scalar weighted case.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:23:09 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cruz-Uribe",
"David",
""
]
] |
2508.13353
|
Lawford Hatcher
|
Lawford Hatcher
|
Hot spots in domains of constant curvature
|
29 pages, 0 figures
| null | null | null |
math.SP math.AP math.DG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove constant-curvature analogues of several results regarding the hot spots conjecture in dimension two. Our main theorem shows that the hot spots conjecture holds for all non-acute geodesic triangles of constant negative curvature. We also prove that, under certain circumstances, on constant (positive or negative) curvature triangles, first mixed Dirichlet-Neumann Laplace eigenfunctions have no non-vertex critical points. Moreover, we show that each of these eigenfunctions is monotonic with respect to some Killing field. Finally, we show that for general simply connected polygons of non-zero constant curvature--with exactly one family of exceptions--second Neumann eigenfunctions of the Laplacian have at most finitely many critical points.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:33:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hatcher",
"Lawford",
""
]
] |
2508.13354
|
Paulo Cesar Ventura da Silva
|
Paulo Cesar Ventura, Yong Dam Jeong, Maria Litvinova, Allisandra G. Kummer, Shingo Iwami, Hongjie Yu, Stefano Merler, Alessandro Vespignani, Keisuke Ejima, Marco Ajelli
|
VIBES: A Multi-Scale Modeling Approach Integrating Within-Host and Between-Hosts Dynamics in Epidemics
|
37 pages and 8 figures
| null | null | null |
q-bio.PE
|
http://creativecommons.org/licenses/by/4.0/
|
Infectious disease spread is a multi-scale process composed of within-host (biological) and between-host (social) drivers and disentangling them from each other is a central challenge in epidemiology. Here, we introduce VIBES, a multi-scale modeling framework that explicitly integrates viral dynamics based on patient-level data with population-level transmission on a data-driven network of social contacts. Using SARS-CoV-2 as a case study, we analyze three emergent epidemic properties, namely the generation time, serial interval, and pre-symptomatic transmission. First, we established a purely biological baseline, thus independent of the reproduction number (R), from the within-host model, estimating a generation time of 6.3 days for symptomatic individuals and 43.1% presymptomatic transmission. Then, using the full model incorporating social contacts, we found a shorter generation time (5.4 days at R=3.0) and an increase in pre-symptomatic transmission (52.8% at R=3.0), disentangling the impact of social drivers from a purely biological baseline. We further show that as pathogen transmissibility increases (R from 1.3 to 6), competition among infectious individuals shortens the generation time and serial interval by up to 21% and 13%, respectively. Conversely, a social intervention, like isolation, increases the proportion of pre-symptomatic transmission by about 30%. Our framework also estimates metrics that are challenging to obtain empirically, such as the generation time for asymptomatic individuals (5.6 days; 95%CI: 5.1-6.0 at R=1.3). Our findings establish multi-scale modeling as a powerful tool for mechanistically quantifying how pathogen biology and human social behavior shape epidemic dynamics as well as for assessing public health interventions.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:37:12 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ventura",
"Paulo Cesar",
""
],
[
"Jeong",
"Yong Dam",
""
],
[
"Litvinova",
"Maria",
""
],
[
"Kummer",
"Allisandra G.",
""
],
[
"Iwami",
"Shingo",
""
],
[
"Yu",
"Hongjie",
""
],
[
"Merler",
"Stefano",
""
],
[
"Vespignani",
"Alessandro",
""
],
[
"Ejima",
"Keisuke",
""
],
[
"Ajelli",
"Marco",
""
]
] |
2508.13355
|
Wenhao Mu
|
Wenhao Mu, Zhi Cao, Mehmed Uludag, Alexander Rodr\'iguez
|
Counterfactual Probabilistic Diffusion with Expert Models
| null | null | null | null |
cs.LG cs.AI stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
Predicting counterfactual distributions in complex dynamical systems is essential for scientific modeling and decision-making in domains such as public health and medicine. However, existing methods often rely on point estimates or purely data-driven models, which tend to falter under data scarcity. We propose a time series diffusion-based framework that incorporates guidance from imperfect expert models by extracting high-level signals to serve as structured priors for generative modeling. Our method, ODE-Diff, bridges mechanistic and data-driven approaches, enabling more reliable and interpretable causal inference. We evaluate ODE-Diff across semi-synthetic COVID-19 simulations, synthetic pharmacological dynamics, and real-world case studies, demonstrating that it consistently outperforms strong baselines in both point prediction and distributional accuracy.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:44:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Mu",
"Wenhao",
""
],
[
"Cao",
"Zhi",
""
],
[
"Uludag",
"Mehmed",
""
],
[
"Rodríguez",
"Alexander",
""
]
] |
2508.13356
|
Eliza Cornell
|
Eliza Cornell, Zhujing Xu, Zhaoyou Wang, Hana K. Warner, Eliana Mann, Michael Haas, Smarak Maity, Graham Joe, Liang Jiang, Peter Rabl, Benjamin Pingault, Marko Lon\v{c}ar
|
All-mechanical coherence protection and fast control of a spin qubit
|
17 pages, 8 figures, to be submitted to Nature Physics
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
In a phononic quantum network, quantum information is stored and processed within stationary nodes defined by solid-state spins, and the information is routed between nodes by phonons. The phonon holds distinct advantages over its electromagnetic counterpart the photon, including smaller device footprints, reduced crosstalk, long coherence times at low temperatures, and strong interactions with both solid-state spins and electromagnetic waves. Enhanced interactions between a phononic cavity and a stationary qubit have been demonstrated in multiple platforms including superconducting qubits, spins in silicon carbide and spins in diamond. However, an outstanding issue is the compatibility between the spin's coupling to the resonant phononic cavity and the simultaneous use of pulse sequences to extend the coherence time of the spin by suppressing the low-frequency environmental noise. Here we demonstrate all-mechanical coherence protection of a solid-state spin qubit, where optical initialization, quantum operations, and readout are performed in a dressed basis that is highly immune to low-frequency noise and compatible with a phononic cavities. We additionally show record-high Rabi frequencies reaching 800 MHz, which allows for ultrafast quantum control. Our results establish a first step for high-fidelity, phonon-mediated quantum gates and represent a crucial advance toward robust on-chip quantum phononic networks.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 20:47:44 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cornell",
"Eliza",
""
],
[
"Xu",
"Zhujing",
""
],
[
"Wang",
"Zhaoyou",
""
],
[
"Warner",
"Hana K.",
""
],
[
"Mann",
"Eliana",
""
],
[
"Haas",
"Michael",
""
],
[
"Maity",
"Smarak",
""
],
[
"Joe",
"Graham",
""
],
[
"Jiang",
"Liang",
""
],
[
"Rabl",
"Peter",
""
],
[
"Pingault",
"Benjamin",
""
],
[
"Lončar",
"Marko",
""
]
] |
2508.13357
|
Zhuoran Li
|
Zhuoran Li, Hanieh Totonchi Asl, Ebrahim Nouri, Yifei Cai, Danella Zhao
|
Silentflow: Leveraging Trusted Execution for Resource-Limited MPC via Hardware-Algorithm Co-design
| null | null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Secure Multi-Party Computation (MPC) offers a practical foundation for privacy-preserving machine learning at the edge, with MPC commonly employed to support nonlinear operations. These MPC protocols fundamentally rely on Oblivious Transfer (OT), particularly Correlated OT (COT), to generate correlated randomness essential for secure computation. Although COT generation is efficient in conventional two-party settings with resource-rich participants, it becomes a critical bottleneck in real-world inference on resource-constrained devices (e.g., IoT sensors and wearables), due to both communication latency and limited computational capacity. To enable real-time secure inference, we introduce Silentflow, a highly efficient Trusted Execution Environment (TEE)-assisted protocol that eliminates communication in COT generation. We tackle the core performance bottleneck-low computational intensity-through structured algorithmic decomposition: kernel fusion for parallelism, Blocked On-chip eXpansion (BOX) to improve memory access patterns, and vectorized batch operations to maximize memory bandwidth utilization. Through design space exploration, we balance end-to-end latency and resource demands, achieving up to 39.51x speedup over state-of-the-art protocols. By offloading COT computations to a Zynq-7000 SoC, SilentFlow accelerates PPMLaaS inference on the ImageNet dataset under resource constraints, achieving a 4.62x and 3.95x speedup over Cryptflow2 and Cheetah, respectively.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:00:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Li",
"Zhuoran",
""
],
[
"Asl",
"Hanieh Totonchi",
""
],
[
"Nouri",
"Ebrahim",
""
],
[
"Cai",
"Yifei",
""
],
[
"Zhao",
"Danella",
""
]
] |
2508.13358
|
Zeeshan Ahmed
|
Zeeshan Ahmed, Frank Seide, Niko Moritz, Ju Lin, Ruiming Xie, Simone Merello, Zhe Liu and Christian Fuegen
|
Overcoming Latency Bottlenecks in On-Device Speech Translation: A Cascaded Approach with Alignment-Based Streaming MT
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This paper tackles several challenges that arise when integrating Automatic Speech Recognition (ASR) and Machine Translation (MT) for real-time, on-device streaming speech translation. Although state-of-the-art ASR systems based on Recurrent Neural Network Transducers (RNN-T) can perform real-time transcription, achieving streaming translation in real-time remains a significant challenge. To address this issue, we propose a simultaneous translation approach that effectively balances translation quality and latency. We also investigate efficient integration of ASR and MT, leveraging linguistic cues generated by the ASR system to manage context and utilizing efficient beam-search pruning techniques such as time-out and forced finalization to maintain system's real-time factor. We apply our approach to an on-device bilingual conversational speech translation and demonstrate that our techniques outperform baselines in terms of latency and quality. Notably, our technique narrows the quality gap with non-streaming translation systems, paving the way for more accurate and efficient real-time speech translation.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:00:11 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ahmed",
"Zeeshan",
""
],
[
"Seide",
"Frank",
""
],
[
"Moritz",
"Niko",
""
],
[
"Lin",
"Ju",
""
],
[
"Xie",
"Ruiming",
""
],
[
"Merello",
"Simone",
""
],
[
"Liu",
"Zhe",
""
],
[
"Fuegen",
"Christian",
""
]
] |
2508.13359
|
Wang Chen
|
Wang Chen and Arnold X.-X. Yuan
|
Unified Modelling of Infrastructure Asset Performance Deterioration -- a bounded gamma process approach
| null | null | null | null |
stat.AP cs.CE math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Infrastructure asset management systems require a flexible deterioration model that can handle various degradation patterns in a unified way. Owing to its appealing monotonic sample paths, independent increments and mathematical tractability, gamma process has been widely employed as an infrastructure performance deterioration model. This model was recently enhanced by introducing an upper bound to satisfy a practical modelling need that many infrastructure performance deterioration processes are constrained by physical or managerial limits. Several bounded transformed gamma process (BTGP) alternatives had been proposed; however, they lacked due flexibility to characterize different deterioration patterns. This paper proposed a new BTGP model that is deeply grounded upon the traditional regression modelling tradition in infrastructure asset management systems. Qualitative and quantitative comparisons were carried out between the proposed BTGP and a bounded nonstationary gamma process (BNGP) model from both deterioration modelling and asset management decision-making perspectives. An empirical study using the real-world historical bridge condition data was performed to examine the flexibility of the BTGP against the BNGP and six other BTGP alternatives. The results confirmed the flexibility and significance of the proposed BTGP model for infrastructure systems.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:00:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chen",
"Wang",
""
],
[
"Yuan",
"Arnold X. -X.",
""
]
] |
2508.13360
|
Reinhard Maurer
|
Xuexun Lu and Nils Hertl and Reinhard J. Maurer
|
A Haldane-Anderson Hamiltonian Model for Hyperthermal Hydrogen Scattering from a Semiconductor Surface
|
12 pages, 7 figures
| null | null | null |
cond-mat.mtrl-sci physics.chem-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Collisions of atoms and molecules with metal surfaces create electronic excitations in the metal, leading to nonadiabatic energy dissipation, inelastic scattering, and sticking. Mixed quantum-classical molecular dynamics simulation methods, such as molecular dynamics with electronic friction, are able to capture nonadiabatic energy loss during dynamics at metal surfaces. Hydrogen atom scattering on semiconductors, on the other hand, exhibits strong adsorbate-surface energy transfer only when the projectile kinetic energy exceeds the bandgap of the substrate. Electronic friction fails to describe this effect. Here, we report a first-principles parameterization of a simple Haldane-Anderson Hamiltonian model of hydrogen atom gas-surface scattering on $c(2\times8)$Ge(111), for which hyperthermal scattering experiments have been reported. We subsequently perform independent electron surface hopping and Ehrenfest dynamics simulations on this model. Whereas mean-field dynamics yield weak nonadiabatic energy loss that is independent of the initial kinetic energy, independent electron surface hopping simulations qualitatively agree with the experimental observation that nonadiabatic energy dissipation only occurs if the initial kinetic energy exceeds the bandgap of the surface. We further show how nonadiabatic energy loss affects sticking, which is yet to be experimentally measured.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:00:51 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lu",
"Xuexun",
""
],
[
"Hertl",
"Nils",
""
],
[
"Maurer",
"Reinhard J.",
""
]
] |
2508.13361
|
Bekir Can L\"utf\"uo\u{g}lu
|
B. C. L\"utf\"uo\u{g}lu
|
Long-lived quasinormal modes in the Euler-Heisenberg electrodynamics
|
17 pages, 7 figures
| null | null | null |
gr-qc
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Using the precise Leaver method and time-domain integration, we analyze the quasinormal modes and late-time behavior of massive neutral and charged scalar fields in the background of a charged, asymptotically flat black hole in the presence of Euler-Heisenberg nonlinear electrodynamics. We show that as the field mass increases, the damping rate decreases significantly, approaching arbitrarily long-lived states known as quasi-resonances. However, these modes cannot be identified in time-domain profiles due to the dominance of asymptotic tails in this regime, which decay slowly and exhibit oscillations with a power-law envelope. We observe that a larger field charge leads to a significantly higher quality factor, as it increases the oscillation frequency while reducing the damping rate.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:11:37 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lütfüoğlu",
"B. C.",
""
]
] |
2508.13362
|
Ruipu Li
|
Ruipu Li, Daniel Menacho, Alexander Rodr\'iguez
|
Adaptive Conformal Prediction Intervals Over Trajectory Ensembles
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Future trajectories play an important role across domains such as autonomous driving, hurricane forecasting, and epidemic modeling, where practitioners commonly generate ensemble paths by sampling probabilistic models or leveraging multiple autoregressive predictors. While these trajectories reflect inherent uncertainty, they are typically uncalibrated. We propose a unified framework based on conformal prediction that transforms sampled trajectories into calibrated prediction intervals with theoretical coverage guarantees. By introducing a novel online update step and an optimization step that captures inter-step dependencies, our method can produce discontinuous prediction intervals around each trajectory, naturally capture temporal dependencies, and yield sharper, more adaptive uncertainty estimates.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:14:07 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Li",
"Ruipu",
""
],
[
"Menacho",
"Daniel",
""
],
[
"Rodríguez",
"Alexander",
""
]
] |
2508.13363
|
Pegah Varghaei
|
Pegah Varghaei, Kiran Abraham-Aggarwal, Manoj T. Abraham, Arun Ross
|
Automated Assessment of Aesthetic Outcomes in Facial Plastic Surgery
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We introduce a scalable, interpretable computer-vision framework for quantifying aesthetic outcomes of facial plastic surgery using frontal photographs. Our pipeline leverages automated landmark detection, geometric facial symmetry computation, deep-learning-based age estimation, and nasal morphology analysis. To perform this study, we first assemble the largest curated dataset of paired pre- and post-operative facial images to date, encompassing 7,160 photographs from 1,259 patients. This dataset includes a dedicated rhinoplasty-only subset consisting of 732 images from 366 patients, 96.2% of whom showed improvement in at least one of the three nasal measurements with statistically significant group-level change. Among these patients, the greatest statistically significant improvements (p < 0.001) occurred in the alar width to face width ratio (77.0%), nose length to face height ratio (41.5%), and alar width to intercanthal ratio (39.3%). Among the broader frontal-view cohort, comprising 989 rigorously filtered subjects, 71.3% exhibited significant enhancements in global facial symmetry or perceived age (p < 0.01). Importantly, our analysis shows that patient identity remains consistent post-operatively, with True Match Rates of 99.5% and 99.6% at a False Match Rate of 0.01% for the rhinoplasty-specific and general patient cohorts, respectively. Additionally, we analyze inter-practitioner variability in improvement rates. By providing reproducible, quantitative benchmarks and a novel dataset, our pipeline facilitates data-driven surgical planning, patient counseling, and objective outcome evaluation across practices.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:15:11 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Varghaei",
"Pegah",
""
],
[
"Abraham-Aggarwal",
"Kiran",
""
],
[
"Abraham",
"Manoj T.",
""
],
[
"Ross",
"Arun",
""
]
] |
2508.13364
|
Tadeu Freitas
|
Tadeu Freitas, Carlos Novo, In\^es Dutra, Jo\~ao Soares, Manuel Correia, Benham Shariati, Rolando Martins
|
A Risk Manager for Intrusion Tolerant Systems: Enhancing HAL 9000 with New Scoring and Data Sources
| null | null | null | null |
cs.CR cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Intrusion Tolerant Systems (ITSs) have become increasingly critical due to the rise of multi-domain adversaries exploiting diverse attack surfaces. ITS architectures aim to tolerate intrusions, ensuring system compromise is prevented or mitigated even with adversary presence. Existing ITS solutions often employ Risk Managers leveraging public security intelligence to adjust system defenses dynamically against emerging threats. However, these approaches rely heavily on databases like NVD and ExploitDB, which require manual analysis for newly discovered vulnerabilities. This dependency limits the system's responsiveness to rapidly evolving threats. HAL 9000, an ITS Risk Manager introduced in our prior work, addressed these challenges through machine learning. By analyzing descriptions of known vulnerabilities, HAL 9000 predicts and assesses new vulnerabilities automatically. To calculate the risk of a system, it also incorporates the Exploitability Probability Scoring system to estimate the likelihood of exploitation within 30 days, enhancing proactive defense capabilities.
Despite its success, HAL 9000's reliance on NVD and ExploitDB knowledge is a limitation, considering the availability of other sources of information. This extended work introduces a custom-built scraper that continuously mines diverse threat sources, including security advisories, research forums, and real-time exploit proofs-of-concept. This significantly expands HAL 9000's intelligence base, enabling earlier detection and assessment of unverified vulnerabilities. Our evaluation demonstrates that integrating scraper-derived intelligence with HAL 9000's risk management framework substantially improves its ability to address emerging threats. This paper details the scraper's integration into the architecture, its role in providing additional information on new threats, and the effects on HAL 9000's management.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:16:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Freitas",
"Tadeu",
""
],
[
"Novo",
"Carlos",
""
],
[
"Dutra",
"Inês",
""
],
[
"Soares",
"João",
""
],
[
"Correia",
"Manuel",
""
],
[
"Shariati",
"Benham",
""
],
[
"Martins",
"Rolando",
""
]
] |
2508.13365
|
Dylan Phelps
|
Dylan Phelps, Rodrigo Wilkens, Edward Gow-Smith, Thomas Pickard, Maggie Mi, Aline Villavicencio
|
Stands to Reason: Investigating the Effect of Reasoning on Idiomaticity Detection
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
The recent trend towards utilisation of reasoning models has improved the performance of Large Language Models (LLMs) across many tasks which involve logical steps. One linguistic task that could benefit from this framing is idiomaticity detection, as a potentially idiomatic expression must first be understood before it can be disambiguated and serves as a basis for reasoning. In this paper, we explore how reasoning capabilities in LLMs affect idiomaticity detection performance and examine the effect of model size. We evaluate, as open source representative models, the suite of DeepSeek-R1 distillation models ranging from 1.5B to 70B parameters across four idiomaticity detection datasets. We find the effect of reasoning to be smaller and more varied than expected. For smaller models, producing chain-of-thought (CoT) reasoning increases performance from Math-tuned intermediate models, but not to the levels of the base models, whereas larger models (14B, 32B, and 70B) show modest improvements. Our in-depth analyses reveal that larger models demonstrate good understanding of idiomaticity, successfully producing accurate definitions of expressions, while smaller models often fail to output the actual meaning. For this reason, we also experiment with providing definitions in the prompts of smaller models, which we show can improve performance in some cases.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:17:09 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Phelps",
"Dylan",
""
],
[
"Wilkens",
"Rodrigo",
""
],
[
"Gow-Smith",
"Edward",
""
],
[
"Pickard",
"Thomas",
""
],
[
"Mi",
"Maggie",
""
],
[
"Villavicencio",
"Aline",
""
]
] |
2508.13366
|
Aleksei Opacic
|
Aleksei Opacic
|
Monotonic Path-Specific Effects: Application to Estimating Educational Returns
| null | null | null | null |
stat.AP econ.GN q-fin.EC stat.ME
|
http://creativecommons.org/licenses/by/4.0/
|
Conventional research on educational effects typically either employs a "years of schooling" measure of education, or dichotomizes attainment as a point-in-time treatment. Yet, such a conceptualization of education is misaligned with the sequential process by which individuals make educational transitions. In this paper, I propose a causal mediation framework for the study of educational effects on outcomes such as earnings. The framework considers the effect of a given educational transition as operating indirectly, via progression through subsequent transitions, as well as directly, net of these transitions. I demonstrate that the average treatment effect (ATE) of education can be additively decomposed into mutually exclusive components that capture these direct and indirect effects. The decomposition has several special properties which distinguish it from conventional mediation decompositions of the ATE, properties which facilitate less restrictive identification assumptions as well as identification of all causal paths in the decomposition. An analysis of the returns to high school completion in the NLSY97 cohort suggests that the payoff to a high school degree stems overwhelmingly from its direct labor market returns. Mediation via college attendance, completion and graduate school attendance is small because of individuals' low counterfactual progression rates through these subsequent transitions.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:17:11 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Opacic",
"Aleksei",
""
]
] |
2508.13367
|
Stefan Hiemer
|
Stefan Hiemer, Paolo Moretti, Stefano Zapperi, Michael Zaiser
|
Fiber bundle model of thermally activated creep failure
| null | null | null | null |
cond-mat.stat-mech cond-mat.mtrl-sci
|
http://creativecommons.org/licenses/by/4.0/
|
An equal load sharing fiber bundle model for thermally activated breakdown is developed using transition state theory to describe the rate of elementary failures. The lifetime distribution, average, variance and their asymptotic limits for uniform fiber failure thresholds are derived and found to be in excellent agreement with simulations. The asymptotic scaling with regards to the number of fibers matches analytical approximations in the low temperature limit derived by Roux and co-workers for a model of thermal breakdown by stationary Gaussian noise. For the case of randomly distributed fiber failure strengths, the lifetime distribution is derived as a multidimensional integral with no closed form solution. Simulations with different fiber strength distributions indicate that, in the limit of large fiber numbers, the statistics of bundle lifetimes shows a similar asymptotic scaling for distributed and for uniform thresholds. Fiber breakage by thermal activation occurs in avalanches triggered by individual thermally activated failure events, and the asymptotic avalanche size distribution obtained from the simulations matches earlier theoretical results derived for quasistatic loading.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:19:28 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hiemer",
"Stefan",
""
],
[
"Moretti",
"Paolo",
""
],
[
"Zapperi",
"Stefano",
""
],
[
"Zaiser",
"Michael",
""
]
] |
2508.13369
|
Laura Wakelin
|
Kyle Hayden, Lisa Piccirillo, Laura Wakelin
|
Dehn surgery functions are never injective
|
25 pages, 20 figures
| null | null | null |
math.GT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove that, for each fixed rational number $p/q \in \mathbb{Q}$, there exists a pair of distinct knots whose $p/q$-surgeries are orientation-preservingly homeomorphic. This confirms a 1978 conjecture of Gordon.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:20:08 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hayden",
"Kyle",
""
],
[
"Piccirillo",
"Lisa",
""
],
[
"Wakelin",
"Laura",
""
]
] |
2508.13370
|
Amanda Bienz
|
Gerald Collom and Jason Burmark and Olga Pearce and Amanda Bienz
|
Persistent and Partitioned MPI for Stencil Communication
| null | null |
10.1109/HPEC62836.2024.10938452
| null |
cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Many parallel applications rely on iterative stencil operations, whose performance are dominated by communication costs at large scales. Several MPI optimizations, such as persistent and partitioned communication, reduce overheads and improve communication efficiency through amortized setup costs and reduced synchronization of threaded sends. This paper presents the performance of stencil communication in the Comb benchmarking suite when using non blocking, persistent, and partitioned communication routines. The impact of each optimization is analyzed at various scales. Further, the paper presents an analysis of the impact of process count, thread count, and message size on partitioned communication routines. Measured timings show that persistent MPI communication can provide a speedup of up to 37% over the baseline MPI communication, and partitioned MPI communication can provide a speedup of up to 68%.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:20:29 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Collom",
"Gerald",
""
],
[
"Burmark",
"Jason",
""
],
[
"Pearce",
"Olga",
""
],
[
"Bienz",
"Amanda",
""
]
] |
2508.13371
|
Ronit Virwani
|
Ronit Virwani and Ruchika Suryawanshi
|
LOOP: A Plug-and-Play Neuro-Symbolic Framework for Enhancing Planning in Autonomous Systems
|
Submitted to IAAI-26
| null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Planning is one of the most critical tasks in autonomous systems, where even a small error can lead to major failures or million-dollar losses. Current state-of-the-art neural planning approaches struggle with complex domains, producing plans with missing preconditions, inconsistent goals, and hallucinations. While classical planners provide logical guarantees, they lack the flexibility and natural language understanding capabilities needed for modern autonomous systems. Existing neuro-symbolic approaches use one-shot translation from natural language to formal plans, missing the opportunity for neural and symbolic components to work and refine solutions together. To address this gap, we develop LOOP -- a novel neuro-symbolic planning framework that treats planning as an iterative conversation between neural and symbolic components rather than simple translation. LOOP integrates 13 coordinated neural features including graph neural networks for spatial relationships, multi-agent validation for consensus-based correctness, hierarchical decomposition for complex task management, and causal memory that learns from both successes and failures. Unlike existing approaches, LOOP generates PDDL specifications, refines them iteratively based on symbolic feedback, and builds a causal knowledge base from execution traces. LOOP was evaluated on six standard IPC benchmark domains, where it achieved 85.8% success rate compared to LLM+P (55.0%), LLM-as-Planner (19.2%), and Tree-of-Thoughts (3.3%). This work shows that the key to reliable planning is not in choosing between neural networks or symbolic reasoners but it lies in making them actually ``talk'' to each other during the entire process. LOOP provides a thorough blueprint for building autonomous systems that can finally be trusted with critical real-world applications.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:21:21 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Virwani",
"Ronit",
""
],
[
"Suryawanshi",
"Ruchika",
""
]
] |
2508.13372
|
Paul Maguire
|
Nourhan Hendawy, Harold McQuaid, Somhairle Mag Uidhir, David Rutherford, Declan Diver, Davide Mariotti, Paul Maguire
|
Free electron charging of microdroplets in a plasma at atmospheric pressure
|
Main manuscript 7 pages. Supplementary information 15 pages. Submission to
| null | null | null |
physics.plasm-ph
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Gas phase microdroplets have recently demonstrated exceptional chemical properties via suspected mechanisms such as contact electrification and surface charge pinning leading, in turn, to high electric fields at the surface. In this work, microdroplets are injected into a low temperature flowing gas plasma at atmospheric pressure and during flight they are subject to an excess free electron flux. We report the first measurements of particle charge acquired in a fully collisional plasma with average values up to 2.5E5 electrons, dependent on absorbed power, for droplets with an average diameter of 15 um. Simulations indicate that for similar plasma conditions, the acquired charge on solid particles would be ~40% lower, at 1.5 x 105 electrons, with the difference likely due in part to low mobility water cluster ion formation around the evaporating droplet. Simulation also indicates surface electric fields up to 1E7 V m-1 for the smallest droplets (3 um). The formation of H2O2 was observed in the liquid with measured values up to 33 mM, equivalent to a remarkable generation rate of over 275 M s-1. The microdroplet system involves a complex interplay of various potential mechanisms, yet to be elucidated. The inclusion of a controlled plasma environment offers the capability for creating known charge levels and defined exposure times to allow more systematic study as well as enhancing their already exceptional characteristics.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:27:54 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hendawy",
"Nourhan",
""
],
[
"McQuaid",
"Harold",
""
],
[
"Uidhir",
"Somhairle Mag",
""
],
[
"Rutherford",
"David",
""
],
[
"Diver",
"Declan",
""
],
[
"Mariotti",
"Davide",
""
],
[
"Maguire",
"Paul",
""
]
] |
2508.13373
|
Joel Peck
|
Joel R. Peck, David Waxman
|
How can we measure the information created by natural selection?
| null | null | null | null |
q-bio.PE
|
http://creativecommons.org/licenses/by/4.0/
|
Natural selection can create information. In particular, because of the action of natural selection, we can often learn something about an environment by examining local organisms, and vice versa. For example, the characteristics of a cactus suggest that the local environment is relatively dry, and if a natural terrestrial environment is dry, then we will generally have an enhanced probability of finding drought-resistant plants (like cacti). Here, we propose a measure that can be used to quantify the information that is created by natural selection. We call the proposed quantity reproductive information, and we show that it has an intuitively satisfying relationship to standard quantitative definitions of information. Reproductive information is also approximately equal to a previously defined measure of biological adaptation. In addition, we explain how reproductive information can be measured using phenotypic characters, instead of genotypes. This could facilitate the measurement of reproductive information, and it could also allow for the quantification of the information that is created by natural selection on groups of organisms, instead of just selection on individuals. Thus, the concept of reproductive information has the potential to advance research on the "units of selection", the "major transitions in evolution", and the emergence of "superorganisms" via cooperation among group members.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:29:25 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Peck",
"Joel R.",
""
],
[
"Waxman",
"David",
""
]
] |
2508.13374
|
Ruozhou Yu
|
Zhouyu Li, Zhijing Yang, Huayue Gu, Xiaojian Wang, Yuchen Liu, Ruozhou Yu
|
OrbitChain: Orchestrating In-orbit Real-time Analytics of Earth Observation Data
|
currently under review
| null | null | null |
cs.DC cs.ET cs.LG cs.NI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Earth observation analytics have the potential to serve many time-sensitive applications. However, due to limited bandwidth and duration of ground-satellite connections, it takes hours or even days to download and analyze data from existing Earth observation satellites, making real-time demands like timely disaster response impossible. Toward real-time analytics, we introduce OrbitChain, a collaborative analytics framework that orchestrates computational resources across multiple satellites in an Earth observation constellation. OrbitChain decomposes analytics applications into microservices and allocates computational resources for time-constrained analysis. A traffic routing algorithm is devised to minimize the inter-satellite communication overhead. OrbitChain adopts a pipeline workflow that completes Earth observation tasks in real-time, facilitates time-sensitive applications and inter-constellation collaborations such as tip-and-cue. To evaluate OrbitChain, we implement a hardware-in-the-loop orbital computing testbed. Experiments show that our system can complete up to 60% analytics workload than existing Earth observation analytics framework while reducing the communication overhead by up to 72%.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:33:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Li",
"Zhouyu",
""
],
[
"Yang",
"Zhijing",
""
],
[
"Gu",
"Huayue",
""
],
[
"Wang",
"Xiaojian",
""
],
[
"Liu",
"Yuchen",
""
],
[
"Yu",
"Ruozhou",
""
]
] |
2508.13375
|
Yusuf M\"ucahit \c{C}etinkaya
|
Yusuf M\"ucahit \c{C}etinkaya, Tu\u{g}rulcan Elmas
|
State & Geopolitical Censorship on Twitter (X): Detection & Impact Analysis of Withheld Content
| null | null | null | null |
cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
State and geopolitical censorship on Twitter, now X, has been turning into a routine, raising concerns about the boundaries between criminal content and freedom of speech. One such censorship practice, withholding content in a particular state has renewed attention due to Elon Musk's apparent willingness to comply with state demands. In this study, we present the first quantitative analysis of the impact of state censorship by withholding on social media using a dataset in which two prominent patterns emerged: Russian accounts censored in the EU for spreading state-sponsored narratives, and Turkish accounts blocked within Turkey for promoting militant propaganda. We find that censorship has little impact on posting frequency but significantly reduces likes and retweets by 25%, and follower growth by 90%-especially when the censored region aligns with the account's primary audience. Meanwhile, some Russian accounts continue to experience growth as their audience is outside the withholding jurisdictions. We develop a user-level binary classifier with a transformer backbone and temporal aggregation strategies, aiming to predict whether an account is likely to be withheld. Through an ablation study, we find that tweet content is the primary signal in predicting censorship, while tweet metadata and profile features contribute marginally. Our best model achieves an F1 score of 0.73 and an AUC of 0.83. This work informs debates on platform governance, free speech, and digital repression.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:36:14 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Çetinkaya",
"Yusuf Mücahit",
""
],
[
"Elmas",
"Tuğrulcan",
""
]
] |
2508.13376
|
Duygu Altinok
|
Duygu Altinok
|
Whispering Context: Distilling Syntax and Semantics for Long Speech Transcripts
|
Accepted to IEEE ASRU 2025. This is the preprint, all rights reserved for ASRU2025
| null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
ASR systems often struggle with maintaining syntactic and semantic accuracy in long audio transcripts, impacting tasks like Named Entity Recognition (NER), capitalization, and punctuation. We propose a novel approach that enhances ASR by distilling contextual knowledge from LLaMA models into Whisper. Our method uses two strategies: (1) token level distillation with optimal transport to align dimensions and sequence lengths, and (2) representation loss minimization between sentence embeddings of Whisper and LLaMA, blending syntax and semantics. Evaluations on the Spoken Wikipedia dataset, a benchmark with long audios and rich entities demonstrate significant improvements in Word Error Rate (WER), NER, capitalization, and punctuation success. By introducing novel NER metrics and exploring semantics aware ASR, our work highlights the value of integrating linguistic context into transcription, setting a foundation for robust, context-aware ASR in longform speech.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:37:09 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Altinok",
"Duygu",
""
]
] |
2508.13377
|
Sudhi Mathur
|
Sudhi Mathur and Neil J. Cornish
|
MaxWave: Rapid Maximum Likelihood Wavelet Reconstruction of Non-Gaussian features in Gravitational Wave Data
|
13 pages, 11 figures
| null | null | null |
gr-qc astro-ph.IM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Advancements in the sensitivity of gravitational-wave detectors have increased the detection rate of transient astrophysical signals. We improve the existing BayesWave initialization algorithm and present a rapid, low-latency approximate maximum likelihood solution for reconstructing non-Gaussian features. We include three enhancements: (1) using a modified wavelet basis to eliminate redundant inner product calculations; (2) shifting from traditional time-frequency-quality factor (TFQ) wavelet transforms to time-frequency-time extent (TF$\tau$) transforms to optimize wavelet subtractions; and (3) implementing a downsampled heterodyned wavelet transform to accelerate initial calculations. Our model can be used to de-noise long-duration signals, which include the stochastic gravitational-wave background from numerous unresolved sources and continuous wave signals from isolated sources such as rotating neutron stars. Through our model, we can also aid machine learning applications that aim to classify and understand glitches by providing non-whitened, time and frequency domain, Gaussian-noise-free reconstructions of any glitch.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:46:20 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Mathur",
"Sudhi",
""
],
[
"Cornish",
"Neil J.",
""
]
] |
2508.13378
|
Yiting Wang
|
Yiting Wang, Ziwei Wang, Jiachen Zhong, Di Zhu, Weiyi Li
|
Applications of Small Language Models in Medical Imaging Classification with a Focus on Prompt Strategies
|
Under Review
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Large language models (LLMs) have shown remarkable capabilities in natural language processing and multi-modal understanding. However, their high computational cost, limited accessibility, and data privacy concerns hinder their adoption in resource-constrained healthcare environments. This study investigates the performance of small language models (SLMs) in a medical imaging classification task, comparing different models and prompt designs to identify the optimal combination for accuracy and usability. Using the NIH Chest X-ray dataset, we evaluate multiple SLMs on the task of classifying chest X-ray positions (anteroposterior [AP] vs. posteroanterior [PA]) under three prompt strategies: baseline instruction, incremental summary prompts, and correction-based reflective prompts. Our results show that certain SLMs achieve competitive accuracy with well-crafted prompts, suggesting that prompt engineering can substantially enhance SLM performance in healthcare applications without requiring deep AI expertise from end users.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:48:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Yiting",
""
],
[
"Wang",
"Ziwei",
""
],
[
"Zhong",
"Jiachen",
""
],
[
"Zhu",
"Di",
""
],
[
"Li",
"Weiyi",
""
]
] |
2508.13379
|
Abdulrahman Bukhari
|
Abdulrahman Bukhari, Bullo Mamo, Mst Shamima Hossain, Ziliang Zhang, Mohsen Karimi, Daniel Enright, Patricia Manosalva, Hyoseung Kim
|
Low-Cost Sensing and Classification for Early Stress and Disease Detection in Avocado Plants
| null | null | null | null |
eess.SY cs.SY
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
With rising demands for efficient disease and salinity management in agriculture, early detection of plant stressors is crucial, particularly for high-value crops like avocados. This paper presents a comprehensive evaluation of low-cost sensors deployed in the field for early stress and disease detection in avocado plants. Our monitoring system was deployed across 72 plants divided into four treatment categories within a greenhouse environment, with data collected over six months. While leaf temperature and conductivity measurements, widely used metrics for controlled settings, were found unreliable in field conditions due to environmental interference and positioning challenges, leaf spectral measurements produced statistically significant results when combined with our machine learning approach. For soil data analysis, we developed a two-level hierarchical classifier that leverages domain knowledge about treatment characteristics, achieving 75-86\% accuracy across different avocado genotypes and outperforming conventional machine learning approaches by over 20\%. In addition, performance evaluation on an embedded edge device demonstrated the viability of our approach for resource-constrained environments, with reasonable computational efficiency while maintaining high classification accuracy. Our work bridges the gap between theoretical potential and practical application of low-cost sensors in agriculture and offers insights for developing affordable, scalable monitoring systems.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:48:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bukhari",
"Abdulrahman",
""
],
[
"Mamo",
"Bullo",
""
],
[
"Hossain",
"Mst Shamima",
""
],
[
"Zhang",
"Ziliang",
""
],
[
"Karimi",
"Mohsen",
""
],
[
"Enright",
"Daniel",
""
],
[
"Manosalva",
"Patricia",
""
],
[
"Kim",
"Hyoseung",
""
]
] |
2508.13380
|
Seohyeon Cha
|
Seohyeon Cha, Kevin Chan, Gustavo de Veciana, Haris Vikalo
|
Batching-Aware Joint Model Onloading and Offloading for Hierarchical Multi-Task Inference
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-sa/4.0/
|
The growing demand for intelligent services on resource-constrained edge devices has spurred the development of collaborative inference systems that distribute workloads across end devices, edge servers, and the cloud. While most existing frameworks focus on single-task, single-model scenarios, many real-world applications (e.g., autonomous driving and augmented reality) require concurrent execution of diverse tasks including detection, segmentation, and depth estimation. In this work, we propose a unified framework to jointly decide which multi-task models to deploy (onload) at clients and edge servers, and how to route queries across the hierarchy (offload) to maximize overall inference accuracy under memory, compute, and communication constraints. We formulate this as a mixed-integer program and introduce J3O (Joint Optimization of Onloading and Offloading), an alternating algorithm that (i) greedily selects models to onload via Lagrangian-relaxed submodular optimization and (ii) determines optimal offloading via constrained linear programming. We further extend J3O to account for batching at the edge, maintaining scalability under heterogeneous task loads. Experiments show J3O consistently achieves over $97\%$ of the optimal accuracy while incurring less than $15\%$ of the runtime required by the optimal solver across multi-task benchmarks.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:49:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cha",
"Seohyeon",
""
],
[
"Chan",
"Kevin",
""
],
[
"de Veciana",
"Gustavo",
""
],
[
"Vikalo",
"Haris",
""
]
] |
2508.13381
|
Ethan Vislosky
|
Ethan Vislosky, Zachery Brown, Regina Demina, Edmond Chaussidon
|
Beyond the two-point correlation: constraining primordial non-gaussianity with density perturbation moments
|
10 pages, 9 figures, to be submitted to Astronomy & Astrophysics
| null | null | null |
astro-ph.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Constraining primordial non-Gaussianities (PNGs) in the large-scale cosmic structure (LSS) is an important step in understanding properties of the early universe, specifically in distinguishing between different inflationary models. Measuring PNG relies on evaluating the scale-dependent correlations in the density field. New summary statistics beyond the two- and three-point correlation functions in configuration space and their Fourier-space counterparts, the power- and bi-spectrum may provide increased sensitivity. We introduce a new method for extracting the PNG signal imprinted on the LSS by using the first three Gaussian moments of the normalized correlation in density perturbations, evaluated at varying distance scales. We aim to assess this method's sensitivity to local PNG, parameterized by $f_ {\mathrm{NL}}$. We perform spherical convolutions at a range of scales on dark matter halo simulations to measure the scale-dependent correlations in the density field. From these, we compute the first three moments and compare them to a model expectation vector, parameterized to the second power in $f_{\mathrm{NL}}$. Our method provides about 21% improvement in sensitivity to $f_{\mathrm{NL}}$ with respect to using the two point correlation function alone. Notably, we find that the second moment alone carries nearly as much constraining power as the mean, highlighting the potential of higher-order statistics. Given its simplicity and efficiency, this framework is well-suited for application to current and upcoming large-scale surveys such as DESI.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:50:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Vislosky",
"Ethan",
""
],
[
"Brown",
"Zachery",
""
],
[
"Demina",
"Regina",
""
],
[
"Chaussidon",
"Edmond",
""
]
] |
2508.13382
|
Hela Dellagi
|
Ayoub Ben Chaliah and Hela Dellagi
|
Datarus-R1: An Adaptive Multi-Step Reasoning LLM for Automated Data Analysis
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
We present Datarus-R1-14B, a 14 B-parameter open-weights language model fine-tuned from Qwen 2.5-14B-Instruct to act as a virtual data analyst and graduate-level problem solver. Datarus is trained not on isolated question-answer pairs but on full analytical trajectories including reasoning steps, code execution, error traces, self-corrections, and final conclusions, all captured in a ReAct-style notebook format spanning finance, medicine, numerical analysis, and other quantitative domains. Our training pipeline combines (i) a trajectory-centric synthetic data generator that yielded 144 000 tagged notebook episodes, (ii) a dual-reward framework blending a lightweight tag-based structural signal with a Hierarchical Reward Model (HRM) that scores both single-step soundness and end-to-end coherence, and (iii) a memory-optimized implementation of Group Relative Policy Optimization (GRPO) featuring KV-cache reuse, sequential generation, and reference-model sharding. A cosine curriculum smoothly shifts emphasis from structural fidelity to semantic depth, reducing the format collapse and verbosity that often plague RL-aligned LLMs. A central design choice in Datarus is it dual reasoning interface. In agentic mode the model produces ReAct-tagged steps that invoke Python tools to execute real code; in reflection mode it outputs compact Chain-of-Thought (CoT) traces delimited by <think> and <answer> tags. On demanding postgraduate-level problems, Datarus exhibits an "AHA-moment" pattern: it sketches hypotheses, revises them once or twice, and converges avoiding the circular, token-inflating loops common to contemporary systems. Across standard public benchmarks Datarus surpasses similar size models and even reaches the level of larger reasoning models such as QwQ-32B achieving up to 30% higher accuracy on AIME 2024/2025 and LiveCodeBench while emitting 18-49% fewer tokens per solution.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 21:58:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chaliah",
"Ayoub Ben",
""
],
[
"Dellagi",
"Hela",
""
]
] |
2508.13383
|
Val\'erian Jacques-Dumas
|
Val\'erian Jacques-Dumas, Henk A. Dijkstra
|
Quantification of the cascading tipping probability from the AMOC to the Amazon rainforest
| null | null | null | null |
physics.ao-ph physics.comp-ph
|
http://creativecommons.org/licenses/by/4.0/
|
The Amazon rainforest and the AMOC are two key components of the Earth system and may both collapse under climate change. Due to its influence on precipitation patterns, a collapsed AMOC influences the dynamics of the Amazon rainforest. We investigate this effect using a coupled conceptual AMOC-Amazon model. The Amazon model is based on empirical hydrological data controlled by the AMOC strength. The AMOC model and its influence on the Amazon are tuned using a simulated AMOC collapse in the Community Earth System Model (CESM). Since the collapse of both systems is very rare, we study it using a ``rare-event'' algorithm, which samples such events much more efficiently than direct numerical simulation. This algorithm also allows us to track many observables of interest of the coupled model. We find that in the centre of the Amazon basin an AMOC collapse is a necessary condition for the Amazon rainforest to collapse, due to its important drying effect. Moreover, we are able to quantify the importance of the AMOC in this tipping cascade by computing the conditional probability that the collapse of the Amazon rainforest follows that of the AMOC, given that the Amazon rainforest turns into a savannah within $200$ years.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:00:12 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Jacques-Dumas",
"Valérian",
""
],
[
"Dijkstra",
"Henk A.",
""
]
] |
2508.13384
|
Trevor Wooley
|
Trevor D. Wooley
|
Subconvex $L^p$-sets, Weyl's inequality, and equidistribution
|
36 pages
| null | null | null |
math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We examine sets $\mathscr A$ of natural numbers having the property that for some real number $p\in (0,2)$, one has the subconvex bound $$\int_0^1 \Bigl| \sum_{n\in \mathscr A\cap [1,N]}e(n\alpha)\Bigr|^p\, {\rm d}\alpha \ll N^{-1}|\mathscr A\cap [1,N]|^p.$$ We show that exponential sums over such sets satisfy inequalities analogous to Weyl's inequality, and in many circumstances of the same strength as classical versions of Weyl's bound. We also examine equidistribution of polynomials modulo $1$ in which the summands are restricted to these subconvex $L^p$-sets. In addition, we describe applications to problems involving character sums and averages of arithmetic functions.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:01:26 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wooley",
"Trevor D.",
""
]
] |
2508.13386
|
Ty Trusty
|
Ty Trusty, David I.W. Levin, and Danny M. Kaufman
|
Sparse, Geometry- and Material-Aware Bases for Multilevel Elastodynamic Simulation
|
15 pages,22 figures
| null | null | null |
cs.GR
|
http://creativecommons.org/licenses/by/4.0/
|
We present a multi-level elastodynamics timestep solver for accelerating incremental potential contact (IPC) simulations. Our method retains the robustness of gold standard IPC in the face of intricate geometry, complex heterogeneous material distributions and high resolution input data without sacrificing visual fidelity (per-timestep relative displacement error of $\approx1\%$). The success of our method is enabled by a novel, sparse, geometry- and material-aware basis construction method which allows for the use of fast preconditioned conjugate gradient solvers (in place of a sparse direct solver), but without suffering convergence issues due to stiff or heterogeneous materials. The end result is a solver that produces results visually indistinguishable and quantitatively very close to gold-standard IPC methods but up to $13\times$ faster on identical hardware.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:11:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Trusty",
"Ty",
""
],
[
"Levin",
"David I. W.",
""
],
[
"Kaufman",
"Danny M.",
""
]
] |
2508.13387
|
Thye Shan Ng
|
Thye Shan Ng, Caren Soyeon Han, Eun-Jung Holden
|
SPANER: Shared Prompt Aligner for Multimodal Semantic Representation
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in multimodal Parameter-Efficient Fine-Tuning (PEFT) have significantly improved performance on downstream tasks such as few-shot retrieval. However, most existing approaches focus on task-specific gains while neglecting the structure of the multimodal embedding space. As a result, modality-specific representations often remain isolated, limiting cross-modal generalisation. In this work, we introduce Shared Prompt AligNER (SPANER), a modality-agnostic PEFT framework designed to embed inputs from diverse modalities into a unified semantic space. At its core, SPANER employs a shared prompt mechanism that acts as a conceptual anchor, enabling semantically related instances to converge spatially regardless of modality. This shared prompt design is inherently extensible, supporting the seamless integration of additional modalities, such as audio, without altering the core architecture. Through comprehensive experiments across vision-language and audio-visual benchmarks, SPANER demonstrates competitive few-shot retrieval performance while preserving high semantic coherence in the learned embedding space. Our results highlight the importance of aligning embedding structures, rather than merely tuning adapter weights, for scalable multimodal learning.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:20:42 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Ng",
"Thye Shan",
""
],
[
"Han",
"Caren Soyeon",
""
],
[
"Holden",
"Eun-Jung",
""
]
] |
2508.13388
|
Riya Sinha
|
Riya Sinha, Amelia Acker, Hanlin Li
|
Data Work in Memory Institutions: Why and How Information Professionals Use Wikidata
|
27 pages, 1 figure, 2 tables. Accepted to PACM HCI (CSCW 2025)
| null |
10.1145/3757575
| null |
cs.HC
|
http://creativecommons.org/licenses/by/4.0/
|
Wikidata, an open structured database and a sibling project to Wikipedia, has recently become an important platform for information professionals to share structured metadata from their memory institutions, organizations that maintain public knowledge and cultural heritage materials. While studies have investigated why and how peer producers contribute to Wikidata, the institutional motivations and practices of these organizations are less understood. Given Wikidata's potential role in linking and supporting knowledge infrastructures and open data systems, we examined why and how information professionals in memory institutions use Wikidata as part of their organizational workflow. Through interviews with 15 participants, we identified the three archetypal roles of Wikidata users within memory institutions, providers, acquirers, and mutualists, and the different types of contributions that these institutions bring to Wikidata. We then explored potential collaboration opportunities between memory institutions and other volunteers in Wikidata, discussed the value of the data work conducted by these professionals, and examined how and why they track their contributions. Our work contributes to the wider discussions around collaboration and data work in CSCW by (1) studying the motivations and practices of information professionals, their differences from those doing volunteer work, and opportunities for the Wikidata community to promote more collaborative efforts within memory institutions and with other volunteers and (2) drawing attention to the important data work done by memory institutions on Wikidata and pointing out opportunities to support the contributions of information professionals.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:33:21 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Sinha",
"Riya",
""
],
[
"Acker",
"Amelia",
""
],
[
"Li",
"Hanlin",
""
]
] |
2508.13389
|
Oliver Dicks Dr
|
Oliver A. Dicks, Solveig S. Aamlid, Alannah M. Hallas and Joerg Rottler
|
Expanding the search space of high entropy oxides and predicting synthesizability using machine learning interatomic potentials
| null | null | null | null |
cond-mat.mtrl-sci cond-mat.dis-nn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We propose an efficient computational methodology for predicting the synthesizability of high entropy oxides (HEOs) in a large space of possible candidate compounds. HEOs are a growing field with an enormous potential chemical composition space, and yet the discovery of new HEOs is slow and driven by experimental trial-and-error. In this work, we attempt to speed up this process by using a machine learned interatomic potential offering DFT-level accuracy. Our methodology starts by identifying a set of crystal structures and elements for screening, building a large random unit cell of each composition and structure, then relaxing this structure. The most promising candidates are distinguished based on the variance of the individual cation energies, which we introduce as our entropy descriptor, and the enthalpy of mixing, which is used as the enthalpy descriptor. The approach is applied to tetravalent HEOs, and its validity is confirmed by comparison to alternative descriptors and DFT calculations for a set of 7 elements. The search is then extended to a set of 14 elements and three crystal structures, where it successfully identifies the only known stable 4-component HEO in the $\alpha$-PbO$_2$ structure, as well as predicting several new 5-component candidate systems. This approach can straightforwardly be applied to new sets of elements and structures, allowing for the accelerated discovery of new HEOs.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:33:29 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dicks",
"Oliver A.",
""
],
[
"Aamlid",
"Solveig S.",
""
],
[
"Hallas",
"Alannah M.",
""
],
[
"Rottler",
"Joerg",
""
]
] |
2508.13390
|
William Zhang
|
William Zhang, Yiwen Zhu, Yunlei Lu, Mathieu Demarne, Wenjing Wang, Kai Deng, Nutan Sahoo, Katherine Lin, Miso Cilimdzic, Subru Krishnan
|
FLAIR: Feedback Learning for Adaptive Information Retrieval
|
Accepted to CIKM2025
| null | null | null |
cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
Recent advances in Large Language Models (LLMs) have driven the adoption of copilots in complex technical scenarios, underscoring the growing need for specialized information retrieval solutions. In this paper, we introduce FLAIR, a lightweight, feedback learning framework that adapts copilot systems' retrieval strategies by integrating domain-specific expert feedback. FLAIR operates in two stages: an offline phase obtains indicators from (1) user feedback and (2) questions synthesized from documentation, storing these indicators in a decentralized manner. An online phase then employs a two-track ranking mechanism to combine raw similarity scores with the collected indicators. This iterative setup refines retrieval performance for any query. Extensive real-world evaluations of FLAIR demonstrate significant performance gains on both previously seen and unseen queries, surpassing state-of-the-art approaches. The system has been successfully integrated into Copilot DECO, serving thousands of users at Microsoft, demonstrating its scalability and effectiveness in operational environments.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:37:27 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhang",
"William",
""
],
[
"Zhu",
"Yiwen",
""
],
[
"Lu",
"Yunlei",
""
],
[
"Demarne",
"Mathieu",
""
],
[
"Wang",
"Wenjing",
""
],
[
"Deng",
"Kai",
""
],
[
"Sahoo",
"Nutan",
""
],
[
"Lin",
"Katherine",
""
],
[
"Cilimdzic",
"Miso",
""
],
[
"Krishnan",
"Subru",
""
]
] |
2508.13391
|
Niamh O'Neill
|
Niamh O'Neill, Benjamin X. Shi, William Baldwin, William C. Witt, G\'abor Cs\'anyi, Julian D. Gale, Angelos Michaelides, Christoph Schran
|
Towards Routine Condensed Phase Simulations with Delta-Learned Coupled Cluster Accuracy: Application to Liquid Water
| null | null | null | null |
physics.chem-ph cond-mat.mtrl-sci physics.comp-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Simulating liquid water to an accuracy that matches its wealth of available experimental data requires both precise electronic structure methods and reliable sampling of nuclear (quantum) motion. This is challenging because applying the electronic structure method of choice - coupled cluster theory with single, double and perturbative triple excitations [CCSD(T)] - to condensed phase systems is currently limited by its computational cost and complexity. Recent tour-de-force efforts have demonstrated that this accuracy can indeed bring simulated liquid water into close agreement with experiment using machine learning potentials (MLPs). However, achieving this remains far from routine, requiring large datasets and significant computational cost. In this work, we introduce a practical approach that combines developments in MLPs with local correlation approximations to enable routine CCSD(T)-level simulations of liquid water. When combined with nuclear quantum effects, we achieve agreement to experiments for structural and transport properties. Crucially, this approach extends beyond constant volume to constant pressure simulations, allowing fundamental properties such as the density to now be predicted by MLP-based CCSD(T) models. Importantly, the approach also handles constant pressure simulations, enabling MLP-based CCSD(T) models to predict isothermal-isobaric bulk properties, such as water's density maximum in close agreement with experiment. Encompassing tests across electronic structure, datasets and MLP architecture, this work provides a practical blueprint towards routinely developing CCSD(T)-based MLPs for the condensed phase.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:38:53 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"O'Neill",
"Niamh",
""
],
[
"Shi",
"Benjamin X.",
""
],
[
"Baldwin",
"William",
""
],
[
"Witt",
"William C.",
""
],
[
"Csányi",
"Gábor",
""
],
[
"Gale",
"Julian D.",
""
],
[
"Michaelides",
"Angelos",
""
],
[
"Schran",
"Christoph",
""
]
] |
2508.13392
|
Sidharth Talia
|
Sidharth Talia, Oren Salzman, Siddhartha Srinivasa
|
Incremental Generalized Hybrid A*
|
8 pages, 7 figures
| null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We address the problem of efficiently organizing search over very large trees, which arises in many applications ranging from autonomous driving to aerial vehicles. Here, we are motivated by off-road autonomy, where real-time planning is essential. Classical approaches use graphs of motion primitives and exploit dominance to mitigate the curse of dimensionality and prune expansions efficiently. However, for complex dynamics, repeatedly solving two-point boundary-value problems makes graph construction too slow for fast kinodynamic planning. Hybrid A* (HA*) addressed this challenge by searching over a tree of motion primitives and introducing approximate pruning using a grid-based dominance check. However, choosing the grid resolution is difficult: too coarse risks failure, while too fine leads to excessive expansions and slow planning. We propose Incremental Generalized Hybrid A* (IGHA*), an anytime tree-search framework that dynamically organizes vertex expansions without rigid pruning. IGHA* provably matches or outperforms HA*. For both on-road kinematic and off-road kinodynamic planning queries for a car-like robot, variants of IGHA* use 6x fewer expansions to the best solution compared to an optimized version of HA*. In simulated off-road experiments in a high fidelity simulator, IGHA* outperforms HA*M when both are used in the loop with a model predictive controller. We demonstrate real-time performance both in simulation and on a small-scale off-road vehicle, enabling fast, robust planning under complex dynamics. Code: https://github.com/personalrobotics/IGHAStar
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:41:28 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Talia",
"Sidharth",
""
],
[
"Salzman",
"Oren",
""
],
[
"Srinivasa",
"Siddhartha",
""
]
] |
2508.13393
|
Haixin Liu
|
Haixin Liu, Alexa Carollo, Jizhao Zang, Scott B. Papp
|
Next-generation soliton frequency combs in photonic-crystal and nanocomposite microresonators
| null | null | null | null |
physics.optics nlin.PS physics.app-ph
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Microresonator frequency combs offer tremendous opportunity to advance applications in fundamental research and technology by linking the optical and microwave frequency domains. Kerr-nonlinear microresonators further enable the generation of portable, integrated optical frequency combs, which are called microcombs. However, the dispersion engineering usually suffers from the small, geometric parameter space, and achieving soliton microcombs is challenging and usually requires complicated experimental techniques and setups. In recent years, the invention of photonic-crystal resonators (PhCRs) provides access to solitons in a convenient and stable way while its mechanism has not been fully understood. In this article, we highlight the perspectives of generating solitons for various applications in PhCRs and give a thorough understanding of the dynamics of soliton formation. We also propose a nanocomposite waveguide structure for the optimization of group velocity dispersion (GVD), lifting the limitation on geometric parameter space. We apply it to the design of the pump-harmonic microcomb, a new concept for f-2f self-referencing in microcombs. Inspired by the pulse-driven microresonator, we provide a scheme of two-microresonator network, retaining the convenience of using a continuous wave laser for microcomb generation. Our work depicts a blueprint of achieving the next generation soliton frequency combs in PhCR and nanocomposite microresonators and highlights their great prospects in optical metrology, precision measurement and optical data transmission.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 22:41:32 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Liu",
"Haixin",
""
],
[
"Carollo",
"Alexa",
""
],
[
"Zang",
"Jizhao",
""
],
[
"Papp",
"Scott B.",
""
]
] |
2508.13394
|
Lam Do
|
Lam Thanh Do, Linh Van Nguyen, David Fu, Kevin Chen-Chuan Chang
|
CASPER: Concept-integrated Sparse Representation for Scientific Retrieval
|
11 Pages. Code: https://github.com/louisdo/CASPER
| null | null | null |
cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The exponential growth of scientific literature has made it increasingly difficult for researchers to keep up with the literature. In an attempt to alleviate this problem, we propose CASPER, a sparse retrieval model for scientific search that utilizes tokens and keyphrases as representation units (i.e. dimensions in the sparse embedding space), enabling it to represent queries and documents with research concepts and match them at both granular and conceptual levels. To overcome the lack of suitable training data, we propose mining training data by leveraging scholarly references (i.e. signals that capture how research concepts of papers are expressed in different settings), including titles, citation contexts, author-assigned keyphrases, and co-citations. CASPER outperforms strong dense and sparse retrieval baselines on eight scientific retrieval benchmarks. Moreover, we demonstrate that through simple post-processing, CASPER can be effectively used for the keyphrase generation tasks, achieving competitive performance with the established CopyRNN while producing more diverse keyphrases and being nearly four times faster.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:00:57 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Do",
"Lam Thanh",
""
],
[
"Van Nguyen",
"Linh",
""
],
[
"Fu",
"David",
""
],
[
"Chang",
"Kevin Chen-Chuan",
""
]
] |
2508.13395
|
Svetlana Makarova
|
Andres Fernandez Herrero, Svetlana Makarova
|
Moduli of non-pure sheaves and contractions of Hilbert schemes
|
46 pages, 3 figures, comments welcome
| null | null | null |
math.AG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We define certain stacks of rank one sheaves on a smooth projective variety, and show that they admit proper good moduli spaces. We offer several applications to contractions of subschemes inside Hilbert schemes of points. We construct a surgery diagram via non-GIT wall-crossing, and use the interpretation of the surgery as a fine moduli of sheaves to prove instances Kawamata's DK-hypothesis in this setting.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:02:25 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Herrero",
"Andres Fernandez",
""
],
[
"Makarova",
"Svetlana",
""
]
] |
2508.13396
|
Dinesh Eswararaj
|
Dinesh Eswararaj, Ajay Babu Nellipudi and Vandana Kollati
|
A Comparative Study of Delta Parquet, Iceberg, and Hudi for Automotive Data Engineering Use Cases
|
Published in SSRG International Journal of Computer Science and Engineering (IJCSE), July 2025. This is the authors accepted manuscript. The final published version is available
| null |
10.14445/23488387/IJCSE-V12I17P104
| null |
cs.SE
|
http://creativecommons.org/licenses/by/4.0/
|
The automotive industry generates vast amounts of data from sensors, telemetry, diagnostics, and real-time operations. Efficient data engineering is critical to handle challenges of latency, scalability, and consistency. Modern data lakehouse formats Delta Parquet, Apache Iceberg, and Apache Hudi offer features such as ACID transactions, schema enforcement, and real-time ingestion, combining the strengths of data lakes and warehouses to support complex use cases. This study presents a comparative analysis of Delta Parquet, Iceberg, and Hudi using real-world time-series automotive telemetry data with fields such as vehicle ID, timestamp, location, and event metrics. The evaluation considers modeling strategies, partitioning, CDC support, query performance, scalability, data consistency, and ecosystem maturity. Key findings show Delta Parquet provides strong ML readiness and governance, Iceberg delivers high performance for batch analytics and cloud-native workloads, while Hudi is optimized for real-time ingestion and incremental processing. Each format exhibits tradeoffs in query efficiency, time-travel, and update semantics. The study offers insights for selecting or combining formats to support fleet management, predictive maintenance, and route optimization. Using structured datasets and realistic queries, the results provide practical guidance for scaling data pipelines and integrating machine learning models in automotive applications.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:06:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Eswararaj",
"Dinesh",
""
],
[
"Nellipudi",
"Ajay Babu",
""
],
[
"Kollati",
"Vandana",
""
]
] |
2508.13397
|
Michael Adams
|
Michael Adams, Amanda Bienz
|
Optimizing Allreduce Operations for Heterogeneous Architectures with Multiple Processes per GPU
| null | null | null | null |
cs.DC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large inter-GPU all-reduce operations, prevalent throughout deep learning, are bottlenecked by communication costs. Emerging heterogeneous architectures are comprised of complex nodes, often containing $4$ GPUs and dozens to hundreds of CPU cores per node. Parallel applications are typically accelerated on the available GPUs, using only a single CPU core per GPU while the remaining cores sit idle. This paper presents novel optimizations to large GPU-aware all-reduce operations, extending lane-aware reductions to the GPUs, and notably using multiple CPU cores per GPU to accelerate these operations. These multi-CPU-accelerated GPU-aware lane all-reduces yield speedup of up to $2.45$x for large MPI all-reduces across the NVIDIA A100 GPUs of NCSA's Delta supercomputer. Finally, the approach is extended to NVIDIA's and AMD's collective communication libraries, achieving speedup of up to $1.77$x and $1.71$x, respectively, across $2$ state-of-the-art supercomputers.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:12:04 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Adams",
"Michael",
""
],
[
"Bienz",
"Amanda",
""
]
] |
2508.13398
|
Filippo Ferrari
|
Leo Kruglikov, Filippo Ferrari, Vincenzo Savona
|
Chaos and quantum regimes in $n$-photon driven, dissipative bosonic chains
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the steady-state dynamical regimes of boundary-driven, dissipative bosonic chains subjected to $n$-photon drives. Using the truncated Wigner approximation, we explore how multi-photon drives shape the interplay between quantum fluctuations, nonlinear interactions, and dissipative processes in such quantum systems. We identify two main regimes: a chaotic hydrodynamic regime characterized by the restoration of a local $\mathbb{U}(1)$ symmetry, photon saturation due to Kerr nonlinearity, and spatial prethermalization effects; and a non-chaotic resonant nonlinear wave (RNW) regime exhibiting sub-Poissonian photon statistics, persistent $\mathbb{Z}_n$ symmetry, and quantum-driven phase decoherence. Our findings reveal the universal nature of the hydrodynamic regime and highlight the RNW regime's sensitivity to boundary driving conditions, suggesting novel routes for quantum state engineering in driven-dissipative quantum devices. These results are experimentally relevant for state-of-the-art circuit quantum electrodynamics platforms.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:24:01 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Kruglikov",
"Leo",
""
],
[
"Ferrari",
"Filippo",
""
],
[
"Savona",
"Vincenzo",
""
]
] |
2508.13399
|
Ioannis Xiradakis
|
Panagiota Fatourou (1 and 2), Eric Ruppert (3), Ioannis Xiradakis (1 and 2) ((1) FORTH ICS, Greece, (2) University of Crete, Greece, (3) York University, Canada)
|
Concurrent Double-Ended Priority Queues
| null | null | null | null |
cs.DS
|
http://creativecommons.org/licenses/by/4.0/
|
This work provides the first concurrent implementation specifically designed for a double-ended priority queue (DEPQ). We do this by describing a general way to add an ExtractMax operation to any concurrent priority queue that already supports Insert and ExtractMin operations. The construction uses two linearizable single-consumer priority queues to build a linearizable dual-consumer DEPQ (only one process can perform Extract operations at each end). This construction preserves lock-freedom. We then describe how to use a lock-based combining scheme to allow multiple consumers at each end of the DEPQ. To illustrate the technique, we apply it to a list-based priority queue.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:28:38 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fatourou",
"Panagiota",
"",
"1 and 2"
],
[
"Ruppert",
"Eric",
"",
"York University, Canada"
],
[
"Xiradakis",
"Ioannis",
"",
"1 and 2"
]
] |
2508.13400
|
Korosh Aghayar Gharehbagh
|
S.Nohekhan Shishavan, K. Aghayar Gharehbagh, and H. Sedgi Gamichi
|
Enhanced Sensitivity and Noise Resilience in Two-Qubit Quantum Magnetometers
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a novel two-qubit quantum magnetometer Hamiltonian optimized for enhanced sensitivity and noise resilience. Compared to existing models, our formulation offers advantages in accuracy, robustness against noise, and entanglement dynamics. Using analytical methods, we derive the Quantum Fisher Information (QFI) and the Signal-to-Noise Ratio (SNR), highlighting its practical viability for magnetic field sensing. Our approach bridges theoretical insights with real-world applicability. We further analyze the performance of the magnetometer with a different initial entangled state, revealing the benefits of entanglement for sensitivity. A comparative analysis with leading research in the field underscores the advancements offered by our proposed design. Finally, we discuss the limitations of our current study and suggest potential avenues for future research.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:29:18 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Shishavan",
"S. Nohekhan",
""
],
[
"Gharehbagh",
"K. Aghayar",
""
],
[
"Gamichi",
"H. Sedgi",
""
]
] |
2508.13401
|
Andrei Dumitriu
|
Andrei Dumitriu, Florin Miron, Florin Tatui, Radu Tudor Ionescu, Radu Timofte, Aakash Ralhan, Florin-Alexandru Vasluianu, Shenyang Qian, Mitchell Harley, Imran Razzak, Yang Song, Pu Luo, Yumei Li, Cong Xu, Jinming Chai, Kexin Zhang, Licheng Jiao, Lingling Li, Siqi Yu, Chao Zhang, Kehuan Song, Fang Liu, Puhua Chen, Xu Liu, Jin Hu, Jinyang Xu, Biao Liu
|
AIM 2025 Rip Current Segmentation (RipSeg) Challenge Report
|
Challenge report paper from AIM2025 Workshop at ICCVW 2025
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This report presents an overview of the AIM 2025 RipSeg Challenge, a competition designed to advance techniques for automatic rip current segmentation in still images. Rip currents are dangerous, fast-moving flows that pose a major risk to beach safety worldwide, making accurate visual detection an important and underexplored research task. The challenge builds on RipVIS, the largest available rip current dataset, and focuses on single-class instance segmentation, where precise delineation is critical to fully capture the extent of rip currents. The dataset spans diverse locations, rip current types, and camera orientations, providing a realistic and challenging benchmark.
In total, $75$ participants registered for this first edition, resulting in $5$ valid test submissions. Teams were evaluated on a composite score combining $F_1$, $F_2$, $AP_{50}$, and $AP_{[50:95]}$, ensuring robust and application-relevant rankings. The top-performing methods leveraged deep learning architectures, domain adaptation techniques, pretrained models, and domain generalization strategies to improve performance under diverse conditions.
This report outlines the dataset details, competition framework, evaluation metrics, and final results, providing insights into the current state of rip current segmentation. We conclude with a discussion of key challenges, lessons learned from the submissions, and future directions for expanding RipSeg.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:34:56 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Dumitriu",
"Andrei",
""
],
[
"Miron",
"Florin",
""
],
[
"Tatui",
"Florin",
""
],
[
"Ionescu",
"Radu Tudor",
""
],
[
"Timofte",
"Radu",
""
],
[
"Ralhan",
"Aakash",
""
],
[
"Vasluianu",
"Florin-Alexandru",
""
],
[
"Qian",
"Shenyang",
""
],
[
"Harley",
"Mitchell",
""
],
[
"Razzak",
"Imran",
""
],
[
"Song",
"Yang",
""
],
[
"Luo",
"Pu",
""
],
[
"Li",
"Yumei",
""
],
[
"Xu",
"Cong",
""
],
[
"Chai",
"Jinming",
""
],
[
"Zhang",
"Kexin",
""
],
[
"Jiao",
"Licheng",
""
],
[
"Li",
"Lingling",
""
],
[
"Yu",
"Siqi",
""
],
[
"Zhang",
"Chao",
""
],
[
"Song",
"Kehuan",
""
],
[
"Liu",
"Fang",
""
],
[
"Chen",
"Puhua",
""
],
[
"Liu",
"Xu",
""
],
[
"Hu",
"Jin",
""
],
[
"Xu",
"Jinyang",
""
],
[
"Liu",
"Biao",
""
]
] |
2508.13402
|
Hao Fang
|
Hao Fang, Haoyuan Zhao, Jianxin Shi, Miao Zhang, Guanzhen Wu, Yi Ching Chou, Feng Wang, Jiangchuan Liu
|
Robust Live Streaming over LEO Satellite Constellations: Measurement, Analysis, and Handover-Aware Adaptation
|
Accepted by ACM Multimedia 2024
| null | null | null |
cs.MM eess.IV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Live streaming has experienced significant growth recently. Yet this rise in popularity contrasts with the reality that a substantial segment of the global population still lacks Internet access. The emergence of Low Earth orbit Satellite Networks (LSNs), such as SpaceX's Starlink and Amazon's Project Kuiper, presents a promising solution to fill this gap. Nevertheless, our measurement study reveals that existing live streaming platforms may not be able to deliver a smooth viewing experience on LSNs due to frequent satellite handovers, which lead to frequent video rebuffering events. Current state-of-the-art learning-based Adaptive Bitrate (ABR) algorithms, even when trained on LSNs' network traces, fail to manage the abrupt network variations associated with satellite handovers effectively. To address these challenges, for the first time, we introduce Satellite-Aware Rate Adaptation (SARA), a versatile and lightweight middleware that can seamlessly integrate with various ABR algorithms to enhance the performance of live streaming over LSNs. SARA intelligently modulates video playback speed and furnishes ABR algorithms with insights derived from the distinctive network characteristics of LSNs, thereby aiding ABR algorithms in making informed bitrate selections and effectively minimizing rebuffering events that occur during satellite handovers. Our extensive evaluation shows that SARA can effectively reduce the rebuffering time by an average of $39.41\%$ and slightly improve latency by $0.65\%$ while only introducing an overall loss in bitrate by $0.13\%$.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:41:22 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Fang",
"Hao",
""
],
[
"Zhao",
"Haoyuan",
""
],
[
"Shi",
"Jianxin",
""
],
[
"Zhang",
"Miao",
""
],
[
"Wu",
"Guanzhen",
""
],
[
"Chou",
"Yi Ching",
""
],
[
"Wang",
"Feng",
""
],
[
"Liu",
"Jiangchuan",
""
]
] |
2508.13403
|
Victor dos Santos Ronchim
|
Christina Brech and Victor dos Santos Ronchim
|
Isometries of James-Schreier and Lorentz spaces
|
20 pages, 0 figures
| null | null | null |
math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We study the group of surjective linear isometries on certain real Banach sequence spaces using the preservation of extreme points in the closed unit ball. Our main result provides a characterization of the extreme points of the dual unit ball of the James-Schreier space $V_1$. As a consequence, we show that the only isometries on $V_1$ are $\pm Id$. We also obtain a Banach-Stone-type result for Lorentz sequence spaces, analogous to one proved in \cite{CarothersIsometrisLorentz} for Lorentz function spaces.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:47:46 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Brech",
"Christina",
""
],
[
"Ronchim",
"Victor dos Santos",
""
]
] |
2508.13405
|
Chungwei Lin
|
Chungwei Lin, Qi Ding, Yanting Ma
|
Application of Optimal Control to Time-Resolution Protocol for Quantum Sensing
|
17 pages, 8 figures, single column
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Time-resolution protocol of quantum sensing aims to measure the fast temporal variation of an external field and demands a high field sensitivity in a short interrogation time $\tau$. Since any operation that evolves the quantum state takes time and is counted as part of the interrogation, evaluating the performance of time-resolution protocol requires a complete end-to-end description of the measurement process. In particular, the initial state has to be one of the sensor qubit's eigenstates in the absence of external fields, and the final projective measurements must be performed in the same eigenstate basis. Building upon prior works which proposed limits for time-resolved sensing using a quantum sensor, we apply optimal control theory to optimize the time-resolution protocol. Our analysis indicates that there exists a critical interrogation time $T^*$: when $\tau<T^*$ the optimal protocol is purely bang-bang; when $\tau>T^*$ the optimal protocol involves a singular control during the interrogation. In the short-$\tau$ regime, which is relevant to high time resolution, we propose a ``detune protocol'' that involves only smooth control during the entire interrogation. As the discontinuities of control pose the main obstacles to experimental realization, we expect the presented detune protocol to be practically useful. In the long-$\tau$ regime, the optimal protocol closely resembles the Ramsey sequence; protocols based on maximizing Quantum Fisher Information are constructed to highlight the difference between the theoretically optimal and practically implementable measurements. Effective use of the time-resolution protocol requires a setup where the unknown time-domain signal of interest can be identically and repeatedly generated. As a potentially relevant application, we outline the calibration of baseband flux pulse distortion in the control of superconducting qubits.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:54:31 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Lin",
"Chungwei",
""
],
[
"Ding",
"Qi",
""
],
[
"Ma",
"Yanting",
""
]
] |
2508.13406
|
Nooshin Bahador
|
Nooshin Bahador, Milad Lankarany
|
Semi-Supervised Anomaly Detection Pipeline for SOZ Localization Using Ictal-Related Chirp
|
23 pages, 7 figures
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This study presents a quantitative framework for evaluating the spatial concordance between clinically defined seizure onset zones (SOZs) and statistically anomalous channels identified through time-frequency analysis of chirp events. The proposed pipeline employs a two-step methodology: (1) Unsupervised Outlier Detection, where Local Outlier Factor (LOF) analysis with adaptive neighborhood selection identifies anomalous channels based on spectro-temporal features of chirp (Onset frequency, offset frequency, and temporal duration); and (2) Spatial Correlation Analysis, which computes both exact co-occurrence metrics and weighted index similarity, incorporating hemispheric congruence and electrode proximity. Key findings demonstrate that the LOF-based approach (N neighbors=20, contamination=0.2) effectively detects outliers, with index matching (weighted by channel proximity) outperforming exact matching in SOZ localization. Performance metrics (precision, recall, F1) were highest for seizure-free patients (Index Precision mean: 0.903) and those with successful surgical outcomes (Index Precision mean: 0.865), whereas failure cases exhibited lower concordance (Index Precision mean: 0.460). The key takeaway is that chirp-based outlier detection, combined with weighted spatial metrics, provides a complementary method for SOZ localization, particularly in patients with successful surgical outcomes.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 23:54:59 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bahador",
"Nooshin",
""
],
[
"Lankarany",
"Milad",
""
]
] |
2508.13408
|
Kamran Chitsaz
|
Kamran Chitsaz, Roshan Balaji, Quentin Fournier, Nirav Pravinbhai Bhatt, Sarath Chandar
|
NovoMolGen: Rethinking Molecular Language Model Pretraining
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Designing de-novo molecules with desired property profiles requires efficient exploration of the vast chemical space ranging from $10^{23}$ to $10^{60}$ possible synthesizable candidates. While various deep generative models have been developed to design small molecules using diverse input representations, Molecular Large Language Models (Mol-LLMs) based on string representations have emerged as a scalable approach capable of exploring billions of molecules. However, there remains limited understanding regarding how standard language modeling practices such as textual representations, tokenization strategies, model size, and dataset scale impact molecular generation performance. In this work, we systematically investigate these critical aspects by introducing NovoMolGen, a family of transformer-based foundation models pretrained on 1.5 billion molecules for de-novo molecule generation. Through extensive empirical analyses, we identify a weak correlation between performance metrics measured during pretraining and actual downstream performance, revealing important distinctions between molecular and general NLP training dynamics. NovoMolGen establishes new state-of-the-art results, substantially outperforming prior Mol-LLMs and specialized generative models in both unconstrained and goal-directed molecular generation tasks, thus providing a robust foundation for advancing efficient and effective molecular modeling strategies.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:04:48 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Chitsaz",
"Kamran",
""
],
[
"Balaji",
"Roshan",
""
],
[
"Fournier",
"Quentin",
""
],
[
"Bhatt",
"Nirav Pravinbhai",
""
],
[
"Chandar",
"Sarath",
""
]
] |
2508.13409
|
Hamza Hanbali
|
Hamza Hanbali
|
Pricing insurance policies with offsetting relationship
| null |
Annals of Actuarial Science , Volume 16 , Issue 2 , July 2022 , pp. 261 - 287
|
10.1017/S1748499521000208
| null |
econ.TH
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper investigates the benefits of incorporating diversification effects into the pricing process of insurance policies from two different business lines. The paper shows that, for the same risk reduction, insurers pricing policies jointly can have a competitive advantage over those pricing them separately. However, the choice of competitiveness constrains the underwriting flexibility of joint pricers. The paper goes a step further by modeling explicitly the relationship between premiums and the number of customers in each line. Using the total collected premiums as a criterion to compare the competing strategies, the paper provides conditions for the optimal pricing decision based on policyholders' sensitivity to price discounts. The results are illustrated for a portfolio of annuities and assurances. Further, using non-life data from the Brazilian insurance market, an empirical exploration shows that most pairs satisfy the condition for being priced jointly, even when pairwise correlations are high.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:08:10 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hanbali",
"Hamza",
""
]
] |
2508.13410
|
Hui Tan
|
Hui Tan, Pengfei Zhang, Jianmin Yuan, Yongqiang Li
|
$p$-orbital self-organization of ultracold atoms coupled to optical cavities
| null | null | null | null |
cond-mat.quant-gas
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Atoms coupled to optical cavities provide a novel platform for understanding high-orbital exotic phenomena in strongly correlated materials. In this study, we investigate strongly correlated ultracold bosonic gases that are coupled to two orthogonally arranged optical cavities and driven by a blue-detuned running-wave laser field. Our results demonstrate that atoms initially in the $s$-orbital state can be scattered into $p_x$- and $p_y$-orbital states in either a symmetric or asymmetric manner, depending on the frequencies of the two cavities. For the symmetric configuration, we observe that atoms are scattered into the $p_x$- and $p_y$-orbitals equally. In the asymmetric case, photons emitted into one cavity mode suppress the scattering into the orthogonal mode. Notably, the coupling of atoms with multiple cavity modes leads to the emergence of high-orbital self-organized phases, accompanied by orbital-density waves that break different symmetries.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:08:54 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Tan",
"Hui",
""
],
[
"Zhang",
"Pengfei",
""
],
[
"Yuan",
"Jianmin",
""
],
[
"Li",
"Yongqiang",
""
]
] |
2508.13411
|
Huiwen Jia
|
Chuyun Deng, Huiwen Jia
|
Decentralized Contextual Bandits with Network Adaptivity
|
46 Pages, 9 figures
| null | null | null |
cs.LG math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We consider contextual linear bandits over networks, a class of sequential decision-making problems where learning occurs simultaneously across multiple locations and the reward distributions share structural similarities while also exhibiting local differences. While classical contextual bandits assume either fully centralized data or entirely isolated learners, much remains unexplored in networked environments when information is partially shared. In this paper, we address this gap by developing two network-aware Upper Confidence Bound (UCB) algorithms, NetLinUCB and Net-SGD-UCB, which enable adaptive information sharing guided by dynamically updated network weights. Our approach decompose learning into global and local components and as a result allow agents to benefit from shared structure without full synchronization. Both algorithms incur lighter communication costs compared to a fully centralized setting as agents only share computed summaries regarding the homogeneous features. We establish regret bounds showing that our methods reduce the learning complexity associated with the shared structure from $O(N)$ to sublinear $O(\sqrt{N})$, where $N$ is the size of the network. The two algorithms reveal complementary strengths: NetLinUCB excels in low-noise regimes with fine-grained heterogeneity, while Net-SGD-UCB is robust to high-dimensional, high-variance contexts. We further demonstrate the effectiveness of our methods across simulated pricing environments compared to standard benchmarks.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:09:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Deng",
"Chuyun",
""
],
[
"Jia",
"Huiwen",
""
]
] |
2508.13412
|
Vera Delfavero
|
V. Delfavero and S. Ray and H. E. Cook and K. Nathaniel and B. McKernan and K. E. S. Ford and J. Postiglione and E. McPike and R. O'Shaughnessy
|
Prospects for the formation of GW231123 from the AGN channel
| null | null | null | null |
gr-qc astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The recent binary black hole (BBH) merger GW231123 consisted of the merger of two intermediate mass black holes (IMBH) which appear to have large spin magnitudes. Active galactic nuclei (AGN) are very promising environments for IMBH mergers and growth due to high escape velocities. Here we demonstrate how GW231123 can be produced in the AGN channel. Using the McFACTS code, we explore the impact of various choices of the black hole (BH) initial mass function (IMF) on predicted mass and spin magnitudes of BBH mergers from the AGN dynamical formation channel. By integrating the likelihood function for GW231123 with the detectable BBH population predicted from AGN using McFACTS, we demonstrate that GW231123 is consistent with a dynamical BBH merger from the AGN channel. We also postulate that the masses and spin magnitudes of GW231123 are most consistent with a merger of fourth and third generation BHs, for most choices of a segregated BH IMF and AGN lifetime.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:09:27 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Delfavero",
"V.",
""
],
[
"Ray",
"S.",
""
],
[
"Cook",
"H. E.",
""
],
[
"Nathaniel",
"K.",
""
],
[
"McKernan",
"B.",
""
],
[
"Ford",
"K. E. S.",
""
],
[
"Postiglione",
"J.",
""
],
[
"McPike",
"E.",
""
],
[
"O'Shaughnessy",
"R.",
""
]
] |
2508.13413
|
Dennis Brown
|
Dennis Brown, Samuel Mulder
|
Large Language Models as Visualization Agents for Immersive Binary Reverse Engineering
|
Accepted to IEEE VISSOFT 2025
| null | null | null |
cs.HC cs.SE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Immersive virtual reality (VR) offers affordances that may reduce cognitive complexity in binary reverse engineering (RE), enabling embodied and external cognition to augment the RE process through enhancing memory, hypothesis testing, and visual organization. In prior work, we applied a cognitive systems engineering approach to identify an initial set of affordances and implemented a VR environment to support RE through spatial persistence and interactivity. In this work, we extend that platform with an integrated large language model (LLM) agent capable of querying binary analysis tools, answering technical questions, and dynamically generating immersive 3D visualizations in alignment with analyst tasks. We describe the system architecture and our evaluation process and results. Our pilot study shows that while LLMs can generate meaningful 3D call graphs (for small programs) that align with design principles, output quality varies widely. This work raises open questions about the potential for LLMs to function as visualization agents, constructing 3D representations that reflect cognitive design principles without explicit training.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:24:01 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Brown",
"Dennis",
""
],
[
"Mulder",
"Samuel",
""
]
] |
2508.13414
|
Kristina Wicke
|
Charles Semple and Kristina Wicke
|
A sharp lower bound for the number of phylogenetic trees displayed by a tree-child network
| null | null | null | null |
math.CO q-bio.PE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
A normal (phylogenetic) network with $k$ reticulations displays $2^k$ phylogenetic trees. In this paper, we establish an analogous result for tree-child (phylogenetic) networks with no underlying $3$-cycles. In particular, we show that a tree-child network with $k\ge 2$ reticulations and no underlying $3$-cycles displays at least $2^{k/2}$ phylogenetic trees if $k$ is even and at least $\frac{3}{2\sqrt{2}}2^{k/2}$ if $k$ is odd. Moreover, we show that these bounds are sharp and characterise the tree-child networks that attain these bounds.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:25:43 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Semple",
"Charles",
""
],
[
"Wicke",
"Kristina",
""
]
] |
2508.13416
|
Franziska Weber
|
Franziska Weber
|
A convergence proof for a finite element discretization of Chorin's projection method of the incompressible Navier-Stokes equations
| null | null | null | null |
math.NA cs.NA math.AP
|
http://creativecommons.org/licenses/by/4.0/
|
We study Chorin's projection method combined with a finite element spatial discretization for the time-dependent incompressible Navier-Stokes equations. The scheme advances the solution in two steps: a prediction step, which computes an intermediate velocity field that is generally not divergence-free, and a projection step, which enforces incompressibility by projecting this velocity onto the divergence-free subspace. We establish convergence, up to a subsequence, of the numerical approximations generated by this scheme to a Leray-Hopf weak solution of the Navier-Stokes equations, without any additional regularity assumptions beyond square-integrable initial data. A discrete energy inequality yields a priori estimates, which we combine with a new compactness result to prove precompactness of the approximations in $L^2([0,T]\times\Omega)$, where $[0,T]$ is the time interval and $\Omega$ is the spatial domain. Passing to the limit as the discretization parameters vanish, we obtain a weak solution of the Navier-Stokes equations. A central difficulty is that different a priori bounds are available for the intermediate and projected velocity fields; our compactness argument carefully integrates these estimates to complete the convergence proof.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:29:40 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Weber",
"Franziska",
""
]
] |
2508.13417
|
Mark Bandstra
|
Mark S. Bandstra and James M. Ghawaly and Douglas E. Peplow and Daniel E. Archer and Brian J. Quiter and Tenzing H.Y. Joshi and Andrew D. Nicholson and Michael J. Willis and Irakli Garishvili and Andrew J. Rowe and Brandon R. Longmire and Jason T. Nattress
|
Full Spectrum Modeling of In Situ Gamma-ray Detector Measurements with a Focus on Precipitation-Induced Transients
|
27 pages, 9 figures, submitted to the Journal of Environmental Radioactivity
| null | null | null |
physics.ins-det
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Gamma-ray detectors that are deployed outdoors experience increased event rates during precipitation due to the attendant increase in Rn-222 progeny at ground level. The increased radiation due to these decay products (Pb-214 and Bi-214) has been studied for many decades in applications such as atmospheric science and radiation protection. For those applications radon progeny signatures are the signal of interest, while in the fields of radiological and nuclear security and aerial radiological mapping they are a nuisance. When searching for radiological contamination or missing sources, an analyst must take precipitation into account to reduce false alarms, in addition to accounting for static background signatures. To train advanced search algorithms, an effort has been underway to generate synthetic gamma-ray event data that represent a realistic urban area, including occasional rain events to add to the realism. This manuscript describes an effort to analyze and model gamma-ray spectra measured during rainfall by a NaI(Tl) detector located outdoors in order to derive accurate source terms for Pb-214 and Bi-214 at a high frequency (less than 1 minute). All known sources of background were quantitatively modeled across the full gamma-ray spectrum, so that the Pb-214 and Bi-214 activity concentrations on the ground could be inferred from a linear model fit to each spectrum. A physically motivated model was applied to the data to further smooth the fits, which had the benefit of yielding information about the concentrations of the progeny in rainwater and their apparent age, making this the first time full-spectrum modeling has been used for continuous measurements of radon progeny. This approach could lead to studies of radon progeny on shorter timescales than previously possible.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:31:16 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bandstra",
"Mark S.",
""
],
[
"Ghawaly",
"James M.",
""
],
[
"Peplow",
"Douglas E.",
""
],
[
"Archer",
"Daniel E.",
""
],
[
"Quiter",
"Brian J.",
""
],
[
"Joshi",
"Tenzing H. Y.",
""
],
[
"Nicholson",
"Andrew D.",
""
],
[
"Willis",
"Michael J.",
""
],
[
"Garishvili",
"Irakli",
""
],
[
"Rowe",
"Andrew J.",
""
],
[
"Longmire",
"Brandon R.",
""
],
[
"Nattress",
"Jason T.",
""
]
] |
2508.13418
|
Zetai Cen
|
Zetai Cen
|
Identification and Estimation of Multi-order Tensor Factor Models
|
47 pages, 3 figures
| null | null | null |
stat.ME math.ST stat.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We propose a novel framework in high-dimensional factor models to simultaneously analyze multiple tensor time series, each with potentially different tensor orders and dimensionality. The connection between different tensor time series is through their global factors that are correlated to each other. A salient feature of our model is that when all tensor time series have the same order, it can be regarded as an extension of multilevel factor models from vectors to general tensors. Under very mild conditions, we separate the global and local components in the proposed model. Parameter estimation is thoroughly discussed. With strong correlation between global factors and noise allowed, we derive the rates of convergence of our estimators, which can be more superior than those of existing methods for multilevel factor models. We also develop estimators that are more computationally efficient, with rates of convergence spelt out. Extensive experiments are performed under various settings, corroborating with the pronounced theoretical results. As a real application example, we analyze a set of taxi data to study the traffic flow between Times Squares and its neighboring areas.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:31:47 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Cen",
"Zetai",
""
]
] |
2508.13419
|
Ashley Bransgrove
|
Ashley Bransgrove, Andrei M. Beloborodov, Yuri Levin
|
Hyperactive Magnetar Eruptions: Giant Flares, Baryon Ejections, and FRBs
|
8 pages, 4 figures, submitted for publication
| null | null | null |
astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Young neutron stars born with magnetic fields $B\gtrsim 10^{16}$ G become hyperactive as the field inside the star evolves through ambipolar diffusion on a timescale $\sim 10^9$ s. We simulate this process numerically and find that it can eject magnetic loops from the star. The internal magnetic field first diffuses to the crust surrounding the liquid core and then erupts from the surface, taking a significant amount of crustal material with it. The eruption involves magnetic reconnection, generating a giant gamma-ray flare. A significant fraction of the eruption energy is carried by the neutron-rich crustal material, which must go through a phase of decompression and nuclear heating. The massive ejecta should produce additional emission components after the giant flare, including radioactively powered gamma-rays, optical emission, and much later a radio afterglow. The predicted eruptions may rarely happen in observed magnetars in our galaxy, which are relatively old and rarely produce giant flares. The model can, however, explain the extremely powerful flare from SGR 1806-20 in December 2004, its ejecta mass, and afterglow. More active, younger magnetars may produce frequent crustal eruptions and form unusual nebulae. Such hyperactive magnetars are candidates for the central engines of cosmological fast radio bursts (FRBs). We argue that each eruption launches an ultrarelativistic magnetosonic pulse leading the ejecta and steepening into a relativistic shock capable of emitting an FRB.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:35:09 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Bransgrove",
"Ashley",
""
],
[
"Beloborodov",
"Andrei M.",
""
],
[
"Levin",
"Yuri",
""
]
] |
2508.13420
|
Casey Schlortt
|
Anh N. Le, Ronnie Pavlov, and Casey Schlortt
|
On subshifts with low maximal pattern complexity
|
28 pages
| null | null | null |
math.DS
|
http://creativecommons.org/licenses/by/4.0/
|
For a finite alphabet $\mathcal{A}$ and a sequence $x \in \mathcal{A}^{\mathbb{N}}$, Kamae and Zamboni defined the maximal pattern complexity function $p^*_x(n)$ as a natural generalization of usual word complexity. They defined a nonperiodic sequence $x$ to be pattern Sturmian if it achieves the minimal growth rate $p^*_x(n) = 2n$, and asked the question of whether one could classify recurrent pattern Sturmian sequences.
We answer their question by characterizing recurrent pattern Sturmian sequences as one of two known types: either a coding of an irrational circle rotation by two intervals, or an element of what we call a nearly simple Toeplitz subshift.
We also show that nonrecurrent pattern Sturmian sequences are either very close to constant (such examples were given by Kamae and Zamboni) or a (nonrecurrent) coding of an irrational circle rotation by two intervals.
Our main new technique is to use topological properties of the maximal equicontinuous factor (MEF) of the subshift generated by $x$. In this way, we prove a general structural result about sequences with non-superlinear maximal pattern complexity: they are either nonrecurrent or minimal with MEF either an odometer or the product of a circle with a finite cyclic group.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:35:54 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Le",
"Anh N.",
""
],
[
"Pavlov",
"Ronnie",
""
],
[
"Schlortt",
"Casey",
""
]
] |
2508.13421
|
Reuben Rideaux
|
Gabrielle Wehr, Reuben Rideaux, Amaya J. Fox, David R. Lightfoot, Jason Tangen, Jason B. Mattingley, Shane E. Ehrhardt
|
Virtuous Machines: Towards Artificial General Science
| null | null | null | null |
cs.AI cs.ET
|
http://creativecommons.org/licenses/by/4.0/
|
Artificial intelligence systems are transforming scientific discovery by accelerating specific research tasks, from protein structure prediction to materials design, yet remain confined to narrow domains requiring substantial human oversight. The exponential growth of scientific literature and increasing domain specialisation constrain researchers' capacity to synthesise knowledge across disciplines and develop unifying theories, motivating exploration of more general-purpose AI systems for science. Here we show that a domain-agnostic, agentic AI system can independently navigate the scientific workflow - from hypothesis generation through data collection to manuscript preparation. The system autonomously designed and executed three psychological studies on visual working memory, mental rotation, and imagery vividness, executed one new online data collection with 288 participants, developed analysis pipelines through 8-hour+ continuous coding sessions, and produced completed manuscripts. The results demonstrate the capability of AI scientific discovery pipelines to conduct non-trivial research with theoretical reasoning and methodological rigour comparable to experienced researchers, though with limitations in conceptual nuance and theoretical interpretation. This is a step toward embodied AI that can test hypotheses through real-world experiments, accelerating discovery by autonomously exploring regions of scientific space that human cognitive and resource constraints might otherwise leave unexplored. It raises important questions about the nature of scientific understanding and the attribution of scientific credit.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:35:56 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wehr",
"Gabrielle",
""
],
[
"Rideaux",
"Reuben",
""
],
[
"Fox",
"Amaya J.",
""
],
[
"Lightfoot",
"David R.",
""
],
[
"Tangen",
"Jason",
""
],
[
"Mattingley",
"Jason B.",
""
],
[
"Ehrhardt",
"Shane E.",
""
]
] |
2508.13422
|
Hamza Hanbali
|
Hamza Hanbali, Daniel Linders and Jan Dhaene
|
Value-at-Risk, Tail Value-at-Risk and upper tail transform of the sum of two counter-monotonic random variables
| null |
Scandinavian Actuarial Journal, Volume 2023, 2023 - Issue 3
|
10.1080/03461238.2022.2092419
| null |
math.PR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The Value-at-Risk (VaR) of comonotonic sums can be decomposed into marginal VaR's at the same level. This additivity property allows to derive useful decompositions for other risk measures. In particular, the Tail Value-at-Risk (TVaR) and the upper tail transform of comonotonic sums can be written as the sum of their corresponding marginal risk measures. The other extreme dependence situation, involving the sum of two arbitrary counter-monotonic random variables, presents a certain number of challenges. One of them is that it is not straightforward to express the VaR of a counter-monotonic sum in terms of the VaR's of the marginal components of the sum. This paper generalizes the results derived in Chaoubi et al. (2020) by providing decomposition formulas for the VaR, TVaR and the stop-loss transform of the sum of two arbitrary counter-monotonic random variables.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:38:19 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Hanbali",
"Hamza",
""
],
[
"Linders",
"Daniel",
""
],
[
"Dhaene",
"Jan",
""
]
] |
2508.13423
|
Qixin Wang
|
Qixin Wang, Dawei Wang, Kun Chen, Yaowei Hu, Puneet Girdhar, Ruoteng Wang, Aadesh Gupta, Chaitanya Devella, Wenlai Guo, Shangwen Huang, Bachir Aoun, Greg Hayworth, Han Li, Xintao Wu
|
AdaptJobRec: Enhancing Conversational Career Recommendation through an LLM-Powered Agentic System
| null | null | null | null |
cs.IR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
In recent years, recommendation systems have evolved from providing a single list of recommendations to offering a comprehensive suite of topic focused services. To better accomplish this task, conversational recommendation systems (CRS) have progressed from basic retrieval augmented LLM generation to agentic systems with advanced reasoning and self correction capabilities. However, agentic systems come with notable response latency, a longstanding challenge for conversational recommendation systems. To balance the trade off between handling complex queries and minimizing latency, we propose AdaptJobRec, the first conversational job recommendation system that leverages autonomous agent to integrate personalized recommendation algorithm tools. The system employs a user query complexity identification mechanism to minimize response latency. For straightforward queries, the agent directly selects the appropriate tool for rapid responses. For complex queries, the agent uses the memory processing module to filter chat history for relevant content, then passes the results to the intelligent task decomposition planner, and finally executes the tasks using personalized recommendation tools. Evaluation on Walmart's real world career recommendation scenarios demonstrates that AdaptJobRec reduces average response latency by up to 53.3% compared to competitive baselines, while significantly improving recommendation accuracy.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:44:25 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Wang",
"Qixin",
""
],
[
"Wang",
"Dawei",
""
],
[
"Chen",
"Kun",
""
],
[
"Hu",
"Yaowei",
""
],
[
"Girdhar",
"Puneet",
""
],
[
"Wang",
"Ruoteng",
""
],
[
"Gupta",
"Aadesh",
""
],
[
"Devella",
"Chaitanya",
""
],
[
"Guo",
"Wenlai",
""
],
[
"Huang",
"Shangwen",
""
],
[
"Aoun",
"Bachir",
""
],
[
"Hayworth",
"Greg",
""
],
[
"Li",
"Han",
""
],
[
"Wu",
"Xintao",
""
]
] |
2508.13424
|
C\'esar Hern\'andez-Cruz
|
J\'ulio Ara\'ujo, C\'esar Hern\'andez-Cruz, Cl\'audia Linhares
|
Maya-Tupi graphs: a generalization of split graphs
|
26 pages, 16 figures
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
We define the family of Maya-Tupi graphs as those graphs that admit a partition $(A,B)$ of their vertex sets such that $A$ induces a complete multipartite graph where each part has size at most two, and $B$ induces a graph where every connected component is $K_1$ or $K_2$. The family of Maya-Tupi graphs is self complementary, generalizes split graphs, falls into the sparse-dense partitioning schema and is characterized by finitely many forbidden induced subgraphs. Unfortunately, our computational experiments show that the number of minimal forbidden induced subgraphs to characterize Maya-Tupi graphs is greater than 2000.
In this work, we study Maya-Tupi graphs when restricted to some well-known graph classes. We find characterizations in terms of minimal forbidden induced subgraphs for disconnected graphs, trees and cographs; our results imply linear time certifying recognition algorithms for Maya-Tupi graphs within these classes. We also show that Maya-Tupi graphs can be recognized in $\mathcal{O}(n^3)$-time in $C_4$-free graphs and in graphs with bounded neighborhood diversity.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:54:03 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Araújo",
"Júlio",
""
],
[
"Hernández-Cruz",
"César",
""
],
[
"Linhares",
"Cláudia",
""
]
] |
2508.13425
|
Mohamed Elmahallawy
|
Mohamed Elmahallawy, Tie Luo
|
When Secure Aggregation Falls Short: Achieving Long-Term Privacy in Asynchronous Federated Learning for LEO Satellite Networks
| null | null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by/4.0/
|
Secure aggregation is a common technique in federated learning (FL) for protecting data privacy from both curious internal entities (clients or server) and external adversaries (eavesdroppers). However, in dynamic and resource-constrained environments such as low Earth orbit (LEO) satellite networks, traditional secure aggregation methods fall short in two aspects: (1) they assume continuous client availability while LEO satellite visibility is intermittent and irregular; (2) they consider privacy in each communication round but have overlooked the possible privacy leakage through multiple rounds. To address these limitations, we propose LTP-FLEO, an asynchronous FL framework that preserves long-term privacy (LTP) for LEO satellite networks. LTP-FLEO introduces (i) privacy-aware satellite partitioning, which groups satellites based on their predictable visibility to the server and enforces joint participation; (ii) model age balancing, which mitigates the adverse impact of stale model updates; and (iii) fair global aggregation, which treats satellites of different visibility durations in an equitable manner. Theoretical analysis and empirical validation demonstrate that LTP-FLEO effectively safeguards both model and data privacy across multi-round training, promotes fairness in line with satellite contributions, accelerates global convergence, and achieves competitive model accuracy.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:55:04 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Elmahallawy",
"Mohamed",
""
],
[
"Luo",
"Tie",
""
]
] |
2508.13426
|
Kabir Manandhar Shrestha Mr
|
Chunhua Liu, Kabir Manandhar Shrestha, Sukai Huang
|
ALIGN: Word Association Learning for Cross-Cultural Generalization in Large Language Models
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
As large language models (LLMs) increasingly mediate cross-cultural communication, their behavior still reflects the distributional bias of the languages and viewpoints that are over-represented in their pre-training corpora. Yet, it remains a challenge to model and align culture due to limited cultural knowledge and a lack of exploration into effective learning approaches. We introduce a cost-efficient, cognitively grounded remedy: parameter-efficient fine-tuning on native speakers' free word-association norms, which encode implicit cultural schemas. Leveraging English-US and Mandarin associations from the Small-World-of-Words project, we adapt Llama-3.1-8B and Qwen-2.5-7B via supervised fine-tuning (SFT) and PPO-based preference optimization. SFT boosts held-out association Precision at 5 by 16-20% in English and 43-165% in Mandarin, lifts median concreteness by +0.20, and attains human-level valence and arousal. These lexical gains transfer: on World-Values-Survey questions, fine-tuned models shift answer distributions toward the target culture, and on a 50-item high-tension subset, Qwen's Chinese-aligned responses double while Llama's US bias drops by one-third. Our 7-8B models rival or beat vanilla 70B baselines, showing that a few million culture-grounded associations can instill value alignment without costly retraining. Our work highlights both the promise and the need for future research grounded in human cognition in improving cultural alignment in AI models.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 00:55:20 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Liu",
"Chunhua",
""
],
[
"Shrestha",
"Kabir Manandhar",
""
],
[
"Huang",
"Sukai",
""
]
] |
2508.13427
|
Yichi Zhang
|
Yichi Zhang, Forrest W. Crawford
|
Time-varying confounding in epidemic intervention evaluations
| null | null | null | null |
stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
Estimating the causal effect of a time-varying public health intervention on the course of an infectious disease epidemic is an important methodological challenge. During the COVID-19 pandemic, researchers attempted to estimate the effects of social distancing policies, stay-at-home orders, school closures, mask mandates, vaccination programs, and many other interventions on population-level infection outcomes. However, measuring the effect of these interventions is complicated by time-varying confounding: public health interventions are causal consequences of prior outcomes and interventions, as well as causes of future outcomes and interventions. Researchers have shown repeatedly that neglecting time-varying confounding for individual-level longitudinal interventions can result in profoundly biased estimates of causal effects. However, the issue with time-varying confounding bias has often been overlooked in population-level epidemic intervention evaluations. In this paper, we explain why associational modeling to estimate the effects of interventions on epidemic outcomes based on observations can be prone to time-varying confounding bias. Using causal reasoning and model-based simulation, we show how directional bias due to time-varying confounding arises in associational modeling and the misleading conclusions it induces.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 01:02:15 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhang",
"Yichi",
""
],
[
"Crawford",
"Forrest W.",
""
]
] |
2508.13428
|
Basura Fernando
|
Hao Zhang and Chen Li and Basura Fernando
|
Mitigating Easy Option Bias in Multiple-Choice Question Answering
|
Under review
| null | null | null |
cs.CV cs.AI cs.MM
|
http://creativecommons.org/licenses/by/4.0/
|
In this early study, we observe an Easy-Options Bias (EOB) issue in some multiple-choice Visual Question Answering (VQA) benchmarks such as MMStar, RealWorldQA, SEED-Bench, Next-QA, STAR benchmark and Video-MME. This bias allows vision-language models (VLMs) to select the correct answer using only the vision (V) and options (O) as inputs, without the need for the question (Q). Through grounding experiments, we attribute the bias to an imbalance in visual relevance: the correct answer typically aligns more closely with the visual contents than the negative options in feature space, creating a shortcut for VLMs to infer the answer via simply vision-option similarity matching. To fix this, we introduce GroundAttack, a toolkit that automatically generates hard negative options as visually plausible as the correct answer. We apply it to the NExT-QA and MMStar datasets, creating new EOB-free annotations. On these EOB-free annotations, current VLMs approach to random accuracies under (V+O) settings, and drop to non-saturated accuracies under (V+Q+O) settings, providing a more realistic evaluation of VLMs' QA ability. Codes and new annotations will be released soon.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 01:03:45 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Zhang",
"Hao",
""
],
[
"Li",
"Chen",
""
],
[
"Fernando",
"Basura",
""
]
] |
2508.13429
|
Paulo Andr\'e Lima De Castro
|
Paulo Andr\'e Lima de Castro
|
AlphaX: An AI-Based Value Investing Strategy for the Brazilian Stock Market
|
6 pages
| null | null | null |
q-fin.CP cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Autonomous trading strategies have been a subject of research within the field of artificial intelligence (AI) for aconsiderable period. Various AI techniques have been explored to develop autonomous agents capable of trading financial assets. These approaches encompass traditional methods such as neural networks, fuzzy logic, and reinforcement learning, as well as more recent advancements, including deep neural networks and deep reinforcement learning. Many developers report success in creating strategies that exhibit strong performance during simulations using historical price data, a process commonly referred to as backtesting. However, when these strategies are deployed in real markets, their performance often deteriorates, particularly in terms of risk-adjusted returns. In this study, we propose an AI-based strategy inspired by a classical investment paradigm: Value Investing. Financial AI models are highly susceptible to lookahead bias and other forms of bias that can significantly inflate performance in backtesting compared to live trading conditions. To address this issue, we conducted a series of computational simulations while controlling for these biases, thereby reducing the risk of overfitting. Our results indicate that the proposed approach outperforms major Brazilian market benchmarks. Moreover, the strategy, named AlphaX, demonstrated superior performance relative to widely used technical indicators such as the Relative Strength Index (RSI) and Money Flow Index (MFI), with statistically significant results. Finally, we discuss several open challenges and highlight emerging technologies in qualitative analysis that may contribute to the development of a comprehensive AI-based Value Investing framework in the future
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 01:04:38 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"de Castro",
"Paulo André Lima",
""
]
] |
2508.13430
|
Kunihiko Taira
|
Kunihiko Taira, Georgios Rigas, Kai Fukami
|
A critical assessment of machine learning in fluid dynamics
|
10 pages
| null | null | null |
physics.flu-dyn
|
http://creativecommons.org/licenses/by/4.0/
|
The fluid dynamics community has increasingly adopted machine learning to analyze, model, predict, and control a wide range of flows. These methods offer powerful computational capabilities for regression, compression, and optimization. In some cases, machine learning has even outperformed traditional approaches. However, many fluid mechanics problems remain beyond the reach of current machine learning techniques. As the field moves from its current state toward a more mature paradigm, this article offers a critical assessment of the key challenges that must be addressed. Tackling these technical issues will not only deepen our understanding of flow physics but also expand the applicability of machine learning beyond fundamental research. We also highlight the importance of community-maintained datasets and open-source code repositories to accelerate progress in this area. Furthermore, the future success of machine learning in fluid dynamics will depend on effective training -- not only for the next generation of researchers but also for established fluid mechanicians adapting to this evolving landscape. Data-driven fluid dynamics is in its critical transitional state over the next few years to shape its future. This perspective article aims to spark discussions and encourage collaborative efforts to advance the integration of machine learning in fluid dynamics.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 01:06:05 GMT"
}
] | 2025-08-20T00:00:00 |
[
[
"Taira",
"Kunihiko",
""
],
[
"Rigas",
"Georgios",
""
],
[
"Fukami",
"Kai",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.