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
|
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2508.13459
|
Shubham Singh
|
Rohan Chandra, Shubham Singh, Abhishek Jha, Dannon Andrade, Hriday Sainathuni, Katia Sycara
|
Multi-Robot Navigation in Social Mini-Games: Definitions, Taxonomy, and Algorithms
| null | null | null | null |
cs.RO cs.MA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The ``Last Mile Challenge'' has long been considered an important, yet unsolved, challenge for autonomous vehicles, public service robots, and delivery robots. A central issue in this challenge is the ability of robots to navigate constrained and cluttered environments that have high agency (e.g., doorways, hallways, corridor intersections), often while competing for space with other robots and humans. We refer to these environments as ``Social Mini-Games'' (SMGs). Traditional navigation approaches designed for MRN do not perform well in SMGs, which has led to focused research on dedicated SMG solvers. However, publications on SMG navigation research make different assumptions (on centralized versus decentralized, observability, communication, cooperation, etc.), and have different objective functions (safety versus liveness). These assumptions and objectives are sometimes implicitly assumed or described informally. This makes it difficult to establish appropriate baselines for comparison in research papers, as well as making it difficult for practitioners to find the papers relevant to their concrete application. Such ad-hoc representation of the field also presents a barrier to new researchers wanting to start research in this area. SMG navigation research requires its own taxonomy, definitions, and evaluation protocols to guide effective research moving forward. This survey is the first to catalog SMG solvers using a well-defined and unified taxonomy and to classify existing methods accordingly. It also discusses the essential properties of SMG solvers, defines what SMGs are and how they appear in practice, outlines how to evaluate SMG solvers, and highlights the differences between SMG solvers and general navigation systems. The survey concludes with an overview of future directions and open challenges in the field.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 02:33:15 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 02:15:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chandra",
"Rohan",
""
],
[
"Singh",
"Shubham",
""
],
[
"Jha",
"Abhishek",
""
],
[
"Andrade",
"Dannon",
""
],
[
"Sainathuni",
"Hriday",
""
],
[
"Sycara",
"Katia",
""
]
] |
2508.13477
|
Shang Zhu
|
Shang Zhu, Yiwen Ma, Jiangbo He, Xiaozhou Yang, Zhongmou Jia, Min Wei, Yiping Jiao, Jiezhong He, Enna Zhuo, Xuewei Cao, Bingbing Tong, Ziwei Dou, Peiling Li, Jie Shen, Xiaohui Song, Zhaozheng Lyu, Guangtong Liu, Dong Pan, Jianhua Zhao, Bo Lu, Li Lu and Fanming Qu
|
Josephson diode effect in nanowire-based Andreev molecules
|
24 pages, 13 figures
|
Communications Physics 8, 330 (2025)
|
10.1038/s42005-025-02237-4
| null |
cond-mat.mes-hall
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Superconducting systems exhibit non-reciprocal current transport under certain conditions of symmetry breaking, a phenomenon known as the superconducting diode effect. This effect allows for perfect rectification of supercurrent, and has received considerable research interest. We report the observation of the Josephson diode effect (JDE) in nanowire-based Andreev molecules, where the time-reversal and spatial-inversion symmetries of a Josephson junction (JJ) can be nonlocally broken by coherently coupling to another JJ. The JDE can be controlled using both non-local phase and gate voltages. Notably, the non-local phase can induce a sign reversal of the diode efficiency, a manifestation of regulating the probabilities of double elastic cotunneling and double-crossed Andreev reflection. Additionally, the diode efficiency can be further modulated by local and non-local gate voltages, exhibiting a central-peak feature in the gate-voltage space. Our theoretical calculations of the energy spectrum and the Josephson currents align well with the experimental results. These results demonstrate the non-local regulation of the JDE in Andreev molecules, offering significant implications for the control of multi-JJ devices and the development of advanced superconducting devices.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 03:15:33 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 05:36:39 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhu",
"Shang",
""
],
[
"Ma",
"Yiwen",
""
],
[
"He",
"Jiangbo",
""
],
[
"Yang",
"Xiaozhou",
""
],
[
"Jia",
"Zhongmou",
""
],
[
"Wei",
"Min",
""
],
[
"Jiao",
"Yiping",
""
],
[
"He",
"Jiezhong",
""
],
[
"Zhuo",
"Enna",
""
],
[
"Cao",
"Xuewei",
""
],
[
"Tong",
"Bingbing",
""
],
[
"Dou",
"Ziwei",
""
],
[
"Li",
"Peiling",
""
],
[
"Shen",
"Jie",
""
],
[
"Song",
"Xiaohui",
""
],
[
"Lyu",
"Zhaozheng",
""
],
[
"Liu",
"Guangtong",
""
],
[
"Pan",
"Dong",
""
],
[
"Zhao",
"Jianhua",
""
],
[
"Lu",
"Bo",
""
],
[
"Lu",
"Li",
""
],
[
"Qu",
"Fanming",
""
]
] |
2508.13496
|
Jingfan Xia
|
Jingfan Xia, Zhenwei Lin, Qi Deng
|
Revisiting Randomized Smoothing: Nonsmooth Nonconvex Optimization Beyond Global Lipschitz Continuity
| null | null | null | null |
math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
Randomized smoothing is a widely adopted technique for optimizing nonsmooth objective functions. However, its efficiency analysis typically relies on global Lipschitz continuity, a condition rarely met in practical applications. To address this limitation, we introduce a new subgradient growth condition that naturally encompasses a wide range of locally Lipschitz functions, with the classical global Lipschitz function as a special case. Under this milder condition, we prove that randomized smoothing yields a differentiable function that satisfies certain generalized smoothness properties. To optimize such functions, we propose novel randomized smoothing gradient algorithms that, with high probability, converge to $(\delta, \epsilon)$-Goldstein stationary points and achieve a sample complexity of $\tilde{\mathcal{O}}(d^{5/2}\delta^{-1}\epsilon^{-4})$. By incorporating variance reduction techniques, we further improve the sample complexity to $\tilde{\mathcal{O}}(d^{3/2}\delta^{-1}\epsilon^{-3})$, matching the optimal $\epsilon$-bound under the global Lipschitz assumption, up to a logarithmic factor. Experimental results validate the effectiveness of our proposed algorithms.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 04:10:38 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:00:13 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Xia",
"Jingfan",
""
],
[
"Lin",
"Zhenwei",
""
],
[
"Deng",
"Qi",
""
]
] |
2508.13515
|
Yiang Shi
|
Yiang Shi, Xiaoyang Guo, Wei Yin, Mingkai Jia, Qian Zhang, Xiaolin Hu, Wenyu Liu, Xinggang Wang
|
2D Gaussians Meet Visual Tokenizer
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The image tokenizer is a critical component in AR image generation, as it determines how rich and structured visual content is encoded into compact representations. Existing quantization-based tokenizers such as VQ-GAN primarily focus on appearance features like texture and color, often neglecting geometric structures due to their patch-based design. In this work, we explored how to incorporate more visual information into the tokenizer and proposed a new framework named Visual Gaussian Quantization (VGQ), a novel tokenizer paradigm that explicitly enhances structural modeling by integrating 2D Gaussians into traditional visual codebook quantization frameworks. Our approach addresses the inherent limitations of naive quantization methods such as VQ-GAN, which struggle to model structured visual information due to their patch-based design and emphasis on texture and color. In contrast, VGQ encodes image latents as 2D Gaussian distributions, effectively capturing geometric and spatial structures by directly modeling structure-related parameters such as position, rotation and scale. We further demonstrate that increasing the density of 2D Gaussians within the tokens leads to significant gains in reconstruction fidelity, providing a flexible trade-off between token efficiency and visual richness. On the ImageNet 256x256 benchmark, VGQ achieves strong reconstruction quality with an rFID score of 1.00. Furthermore, by increasing the density of 2D Gaussians within the tokens, VGQ gains a significant boost in reconstruction capability and achieves a state-of-the-art reconstruction rFID score of 0.556 and a PSNR of 24.93, substantially outperforming existing methods. Codes will be released soon.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 05:04:10 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 01:19:13 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Shi",
"Yiang",
""
],
[
"Guo",
"Xiaoyang",
""
],
[
"Yin",
"Wei",
""
],
[
"Jia",
"Mingkai",
""
],
[
"Zhang",
"Qian",
""
],
[
"Hu",
"Xiaolin",
""
],
[
"Liu",
"Wenyu",
""
],
[
"Wang",
"Xinggang",
""
]
] |
2508.13516
|
Yueh-Po Peng
|
Yueh-Po Peng, Ting-Kang Wang, Li Su, Vincent K.M. Cheung
|
Is Transfer Learning Necessary for Violin Transcription?
|
Accepted at ISMIR 2025 as Late-Breaking Demo (LBD)
| null | null | null |
cs.SD eess.AS
|
http://creativecommons.org/licenses/by/4.0/
|
Automatic music transcription (AMT) has achieved remarkable progress for instruments such as the piano, largely due to the availability of large-scale, high-quality datasets. In contrast, violin AMT remains underexplored due to limited annotated data. A common approach is to fine-tune pretrained models for other downstream tasks, but the effectiveness of such transfer remains unclear in the presence of timbral and articulatory differences. In this work, we investigate whether training from scratch on a medium-scale violin dataset can match the performance of fine-tuned piano-pretrained models. We adopt a piano transcription architecture without modification and train it on the MOSA dataset, which contains about 30 hours of aligned violin recordings. Our experiments on URMP and Bach10 show that models trained from scratch achieved competitive or even superior performance compared to fine-tuned counterparts. These findings suggest that strong violin AMT is possible without relying on pretrained piano representations, highlighting the importance of instrument-specific data collection and augmentation strategies.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 05:20:24 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:34:51 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Peng",
"Yueh-Po",
""
],
[
"Wang",
"Ting-Kang",
""
],
[
"Su",
"Li",
""
],
[
"Cheung",
"Vincent K. M.",
""
]
] |
2508.13550
|
Anthony Chen
|
Anthony Chen and Robert Krasny
|
A Cubed Sphere Fast Multipole Method
| null | null | null | null |
math.NA cs.NA
|
http://creativecommons.org/licenses/by/4.0/
|
This work describes a new version of the Fast Multipole Method for summing pairwise particle interactions that arise from discretizing integral transforms and convolutions on the sphere. The kernel approximations use barycentric Lagrange interpolation on a quadtree composed of cubed sphere grid cells. The scheme is kernel-independent and requires kernel evaluations only at points on the sphere. Results are presented for the Poisson and biharmonic equations on the sphere, barotropic vorticity equation on a rotating sphere, and self-attraction and loading potential in tidal calculations. A tree code version is also described for comparison, and both schemes are tested in serial and parallel calculations.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 06:18:20 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 04:02:09 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chen",
"Anthony",
""
],
[
"Krasny",
"Robert",
""
]
] |
2508.13554
|
Dmitrii Ostrovskii
|
Dmitrii M. Ostrovskii and Pavel S. Shcherbakov
|
Amplitude maximization in stable systems, Schur positivity, and some conjectures on polynomial interpolation
|
18 pages; minor typo corrections viz. the previous version
| null | null | null |
math.CV cs.SY eess.SY math.CA math.CO math.OC
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For $r > 0$ and integers $t \ge n > 0$, we consider the following problem: maximize the amplitude $|x_t|$ at time $t$, over all complex solutions $x = (x_0, x_1, \dots)$ of arbitrary homogeneous linear difference equations of order $n$ with the characteristic roots in the disc $\{z \in \mathbb{C}: |z| \le r\}$, and with initial values $x_0, \dots, x_{n-1}$ in the unit disc. We find that for any triple $t,n,r$, the maximum is attained with coinciding roots on the boundary circle; in particular, this implies that the peak amplitude $\sup_{t \ge n} |x_t|$ can be maximized explicitly, by studying a unique equation with the characteristic polynomial $(z-r)^n$. Moreover, the optimality of the cophase root configuration holds for origin-centered polydiscs. To prove this result, we first reduce the problem to a certain interpolation problem over monomials, then solve the latter by leveraging the theory of symmetric functions and identifying the associated Schur positivity structure. We also discuss the implications for more general Reinhardt domains. Finally, we study the problem of estimating the derivatives of a real entire function from its values at $n/2$ pairs of complex conjugate points in the unit disc. We propose conjectures on the extremality of the monomial $z^n$, and restate them in terms of Schur polynomials.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 06:23:45 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 03:07:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ostrovskii",
"Dmitrii M.",
""
],
[
"Shcherbakov",
"Pavel S.",
""
]
] |
2508.13569
|
Jihun Kim
|
Jihun Kim
|
Revisiting the Geometrically Decaying Step Size: Linear Convergence for Smooth or Non-Smooth Functions
|
4 pages
| null | null | null |
math.OC
|
http://creativecommons.org/licenses/by/4.0/
|
We revisit the geometrically decaying step size given a positive condition number, under which a locally Lipschitz function shows linear convergence. The positivity does not require the function to satisfy convexity, weak convexity, quasar convexity, or sharpness, but instead amounts to a property strictly weaker than the assumptions used in existing works. We propose a clean and simple subgradient descent algorithm that requires minimal knowledge of problem constants, applicable to either smooth or non-smooth functions.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 07:00:01 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 04:55:12 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kim",
"Jihun",
""
]
] |
2508.13628
|
Ao Chen
|
Ao Chen, Lihe Ding, Tianfan Xue
|
DiffIER: Optimizing Diffusion Models with Iterative Error Reduction
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Diffusion models have demonstrated remarkable capabilities in generating high-quality samples and enhancing performance across diverse domains through Classifier-Free Guidance (CFG). However, the quality of generated samples is highly sensitive to the selection of the guidance weight. In this work, we identify a critical ``training-inference gap'' and we argue that it is the presence of this gap that undermines the performance of conditional generation and renders outputs highly sensitive to the guidance weight. We quantify this gap by measuring the accumulated error during the inference stage and establish a correlation between the selection of guidance weight and minimizing this gap. Furthermore, to mitigate this gap, we propose DiffIER, an optimization-based method for high-quality generation. We demonstrate that the accumulated error can be effectively reduced by an iterative error minimization at each step during inference. By introducing this novel plug-and-play optimization framework, we enable the optimization of errors at every single inference step and enhance generation quality. Empirical results demonstrate that our proposed method outperforms baseline approaches in conditional generation tasks. Furthermore, the method achieves consistent success in text-to-image generation, image super-resolution, and text-to-speech generation, underscoring its versatility and potential for broad applications in future research.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 08:41:49 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:14:16 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chen",
"Ao",
""
],
[
"Ding",
"Lihe",
""
],
[
"Xue",
"Tianfan",
""
]
] |
2508.13654
|
Yuming Huang
|
Rapheal Huang (Yuming), Weilong Guo
|
Input Time Scaling
| null | null | null | null |
cs.LG cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Current Large Language Models (LLMs) are usually post-trained on large-scale carefully curated datasets (data & training scaling) and doing reasoning in test time (inference time scaling). In this work, we present a new scaling paradigm, Input Time Scaling, to complement previous scaling methods by putting resources on queries (input time). During training and testing, we combine meta-knowledge from LLMs to refine inputs with different strategies. We also find a new phenomenon, training-testing co-design there. We need to apply query strategies during both training and testing. Only applying strategies on training or testing would seriously degrade the performance. We are also surprised to find that seemingly low data quality datasets can gain high performance. Adding irrelevant information to the queries, randomly selecting examples from a minimally filtered dataset, can even perform the best. These findings contradict the widely held inductive bias, "garbage in, garbage out". Curating datasets with seemingly high-quality data can even potentially limit the performance ceiling. In addition, models trained on more data with similar quality (15k VS 1k) perform worse, simple dataset size scaling should also be carefully inspected. The good news is that our findings are compatible with the Less is More phenomenon. A small set of examples is enough to evoke high-level reasoning ability. With experiments on models trained on Qwen2.5-32B-Instruct, we are able to reach SOTA performance among 32B models on AIME24(76.7%) and AIME25(76.7%) pass@1. We can further achieve AIME24(76.7%) and AIME25(80%) with a majority vote of three models. Starting from DeepSeek-R1-Distill-Qwen-32B, the best result would be 86.7% on AIME24 and 76.7% on AIME25. To facilitate reproducibility and further research, we are working on open-source our datasets, data pipelines, evaluation results, and checkpoints.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 09:04:13 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 06:41:59 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Huang",
"Rapheal",
"",
"Yuming"
],
[
"Guo",
"Weilong",
""
]
] |
2508.13659
|
Huaian Diao
|
Weisheng Zhou, Huaian Diao, and Hongyu Liu
|
Quasi-Minnaert Resonances in High-contrast acoustic Structures and Applications to Invisibility Cloaking
| null | null | null | null |
physics.optics cs.NA math.AP math.NA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper investigates a novel quasi-Minnaert resonance phenomenon in acoustic wave propagation through high-contrast medium in both two and three dimensions, occurring in the sub-wavelength regime. These media are characterized by physical properties significantly distinct from those of a homogeneous background. The quasi-Minnaert resonance is defined by two primary features: boundary localization, where the $L^2$-norms of the internal total field and the external scattered field exhibit pronounced concentration near the boundary, and surface resonance, marked by highly oscillatory behavior of the fields near the boundary. In contrast to classical Minnaert resonances, which are associated with a discrete spectral spectrum tied to physical parameters, quasi-Minnaert resonances exhibit analogous physical phenomena but with a continuous spectral spectrum. Using layer potential theory and rigorous asymptotic analysis, we demonstrate that the coupling between a high-contrast material structure, particularly with radial geometries, and a carefully designed incident wave is critical for inducing quasi-Minnaert resonances. Extensive numerical experiments, involving radial geometries (e.g., unit disks and spheres) and general-shaped geometries (e.g., hearts, Cassini ovals, and clovers in $\mathbb{R}^2$, and spheres in $\mathbb{R}^3$), validate the occurrence of these resonances. Furthermore, we numerically demonstrate that quasi-Minnaert resonances induce an invisibility cloaking effect in the high-contrast medium. These findings have significant implications for mathematical material science and the development of acoustic cloaking technologies.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 09:06:25 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 12:35:51 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhou",
"Weisheng",
""
],
[
"Diao",
"Huaian",
""
],
[
"Liu",
"Hongyu",
""
]
] |
2508.13666
|
Dangfeng Pan
|
Dangfeng Pan, Zhensu Sun, Cenyuan Zhang, David Lo and Xiaoning Du
|
The Hidden Cost of Readability: How Code Formatting Silently Consumes Your LLM Budget
|
Accepted by ICSE'26 (First Cycle)
| null | null | null |
cs.SE cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Source code is usually formatted with elements like indentation and newlines to improve readability for human developers. However, these visual aids do not seem to be beneficial for large language models (LLMs) in the same way since the code is processed as a linear sequence of tokens. Furthermore, these additional tokens can lead to increased computational costs and longer response times for LLMs. If such formatting elements are non-essential to LLMs, we can reduce such costs by removing them from the code. To figure out the role played by formatting elements, we conduct a comprehensive empirical study to evaluate the impact of code formatting on LLM performance and efficiency. Through large-scale experiments on Fill-in-the-Middle Code Completion tasks across four programming languages (Java, Python, C++, C\#) and ten LLMs-including both commercial and open-source models-we systematically analyze token count and performance when formatting elements are removed. Key findings indicate that LLMs can maintain performance across formatted code and unformatted code, achieving an average input token reduction of 24.5\% with negligible output token reductions. This makes code format removal a practical optimization strategy for improving LLM efficiency. Further exploration reveals that both prompting and fine-tuning LLMs can lead to significant reductions (up to 36.1\%) in output code length without compromising correctness. To facilitate practical applications, we develop a bidirectional code transformation tool for format processing, which can be seamlessly integrated into existing LLM inference workflows, ensuring both human readability and LLM efficiency.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 09:13:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Pan",
"Dangfeng",
""
],
[
"Sun",
"Zhensu",
""
],
[
"Zhang",
"Cenyuan",
""
],
[
"Lo",
"David",
""
],
[
"Du",
"Xiaoning",
""
]
] |
2508.13685
|
Christopher Deninger
|
Christopher Deninger and Deepak Kamlesh
|
A remark on the vanishing of Higgs fields in the $p$-adic Simpson correspondence
| null | null | null |
MPIM-Bonn-2025
|
math.AG math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We give a condition on a $p$-adic representation of the fundamental group of a curve over $\overline{\mathbb{Q}}_p$ which ensures that under the $p$-adic Simpson correspondence the Higgs field vanishes.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 09:38:55 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Deninger",
"Christopher",
""
],
[
"Kamlesh",
"Deepak",
""
]
] |
2508.13687
|
Nathan GORSE
|
Nathan Gorse (INSA Toulouse, IMT), Olivier Roustant (INSA Toulouse, IMT), J\'er\'emy Rohmer (BRGM), D\'eborah Idier (BRGM)
|
Simulation of extreme functionals in meteoceanic data: Application to surge evolution over tidal cycles
| null | null | null | null |
stat.AP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the influence of time-varying meteoceanic conditions on coastal flooding under the prism of rare events. Focusing on conditions observed over half tidal cycles, we observe that such data fall within the framework of functional extreme value theory, but violate standard assumptions due to temporal dependence and short-tailed behavior.a To address this, we propose a two-stage methodology. First, we introduce an autoregressive model to eliminate temporal dependence between cycles. Second, considering the model residuals, we adapt existing techniques based on Pareto processes. This allows us to build a simulator of extreme scenarios, by applying inverse transformations. These simulations depend on an initial time series, which can be randomly selected to tune the desired level of extremes. We validate the simulator performance by comparing simulated times series with observations, through several criteria, based on principal component analysis, extreme value analysis, and classification algorithms. The approach is applied to the surge data, on the G{\^a}vres site, located in southern Brittany, France.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 09:40:16 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:51:21 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gorse",
"Nathan",
"",
"INSA Toulouse, IMT"
],
[
"Roustant",
"Olivier",
"",
"INSA Toulouse, IMT"
],
[
"Rohmer",
"Jérémy",
"",
"BRGM"
],
[
"Idier",
"Déborah",
"",
"BRGM"
]
] |
2508.13727
|
Masateru Ishiguro
|
Sunho Jin, Masateru Ishiguro
|
Simulation of Impact-induced seismic shaking on asteroid (25143) Itokawa to address its resurfacing process
|
12 pages, 11 figures
| null | null | null |
astro-ph.EP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The surface of asteroid (25143) Itokawa shows both fresh and mature terrains, despite its short space weathering timescale of approximately 1000 years, as inferred from recent studies. Seismic shaking triggered by the impact that formed the 8-meter Kamoi crater has been proposed as a possible explanation for the diversity. This study aims to examine whether the seismic shaking induced by the impact could account for the observed spatial variations in space weathering and further constrain the internal structure of Itokawa. Assuming that the Kamoi crater was formed by a recent impact, we conducted three-dimensional seismic wave propagation simulations and applied a simplified landslide model to estimate surface accelerations and boulder displacements. Our results show that even a low-energy case (1 % of the nominal seismic energy) produces surface accelerations sufficient to destabilize the surface materials. The simulated boulder displacements are consistent with the observed distribution of space weathering degrees even on the opposite hemisphere. We estimate the seismic diffusivity to be 1000-2000 m2 s-1 and the seismic efficiency to be in the range of 5.0 x 10-8 to 5.0 x 10-7, implying that Itokawa's interior contains blocks tens of meters across and acts as a strongly scattering medium.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 10:53:05 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Jin",
"Sunho",
""
],
[
"Ishiguro",
"Masateru",
""
]
] |
2508.13767
|
Sergey Sergeyev
|
Hani J. Kbashi, Alberto R. Cuevas, and Sergey Sergeyev
|
Mueller Matrix Polarimetry of Fiber Bragg Grating Strain and Torsion
|
Nine pages, seven figures
| null | null | null |
physics.optics physics.data-an physics.ins-det
|
http://creativecommons.org/licenses/by/4.0/
|
We experimentally demonstrate a polarimetric dual-comb spectroscopy technique for simultaneous strain and torsion sensing using a single-cavity mode-locked fiber laser and fiber Bragg grating (FBG) sensors. Dual-comb generation in a single-cavity fiber laser was achieved by utilizing a piece of high-birefringence fiber and adjusting the in-cavity polarization controller. Fast Fourier Transform analysis was applied to the time-domain Stokes parameters, enabling the detection of FBG spectral shifts induced by strain and torsion. To further enhance discrimination between strain and torsion, we applied a novel approach to extract Mueller matrix elements without using complex adjustable polarization components. We explored the analysis of polarimetric purity of the FBG's Mueller matrix in terms of polarizance, diattenuation, and structural polarization response as a function of FBG strain and torsion. The obtained results enabled the measurement of strain and torsion based on a single FBG, which paves the way for the development of cost-effective shape sensing technologies.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 12:07:21 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 09:03:26 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kbashi",
"Hani J.",
""
],
[
"Cuevas",
"Alberto R.",
""
],
[
"Sergeyev",
"Sergey",
""
]
] |
2508.13778
|
Jaka Pelai\v{c}
|
Jaka Pelai\v{c}
|
The subcritical finite-volume massive sine-Gordon model
|
Fixed an issue with the choice of mollifier (+ typos)
| null | null | null |
math-ph math.MP math.PR
|
http://creativecommons.org/licenses/by/4.0/
|
We present a construction of the finite-volume massive sine-Gordon model in the UV subcritical regime using a renormalization group method. The resulting measure has Gaussian tails, respects toroidal symmetries and is reflection-positive.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 12:26:43 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 15:08:40 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Pelaič",
"Jaka",
""
]
] |
2508.13814
|
Diaa Addeen Abuhani
|
Diaa Addeen Abuhani, Marco Seccaroni, Martina Mazzarello, Imran Zualkernan, Fabio Duarte, Carlo Ratti
|
Unsupervised Urban Tree Biodiversity Mapping from Street-Level Imagery Using Spatially-Aware Visual Clustering
|
27 pages, 7 figures, Nature Format
| null | null | null |
cs.CV cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Urban tree biodiversity is critical for climate resilience, ecological stability, and livability in cities, yet most municipalities lack detailed knowledge of their canopies. Field-based inventories provide reliable estimates of Shannon and Simpson diversity but are costly and time-consuming, while supervised AI methods require labeled data that often fail to generalize across regions. We introduce an unsupervised clustering framework that integrates visual embeddings from street-level imagery with spatial planting patterns to estimate biodiversity without labels. Applied to eight North American cities, the method recovers genus-level diversity patterns with high fidelity, achieving low Wasserstein distances to ground truth for Shannon and Simpson indices and preserving spatial autocorrelation. This scalable, fine-grained approach enables biodiversity mapping in cities lacking detailed inventories and offers a pathway for continuous, low-cost monitoring to support equitable access to greenery and adaptive management of urban ecosystems.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 13:18:22 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 11:25:04 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Abuhani",
"Diaa Addeen",
""
],
[
"Seccaroni",
"Marco",
""
],
[
"Mazzarello",
"Martina",
""
],
[
"Zualkernan",
"Imran",
""
],
[
"Duarte",
"Fabio",
""
],
[
"Ratti",
"Carlo",
""
]
] |
2508.13849
|
Joseph Najnudel
|
Christopher Atherfold, Joseph Najnudel
|
The Fourier coefficients of the critical holomorphic multiplicative chaos
| null | null | null | null |
math.PR math.NT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The holomorphic multiplicative chaos (HMC) is a holomorphic analogue of the Gaussian multiplicative chaos. It arises naturally as the limit in large matrix size of the characteristic polynomial of Haar unitary matrices, and more generally, random matrices following the Circular-$\beta$-Ensemble. In a previous article, Najnudel, Paquette and Simm prove that in the $L^2$ phase $\beta > 4$, the appropriately normalized Fourier coefficient of the HMC converges in distribution to the square root of the total mass of the Gaussian multiplicative chaos on the unit circle, multiplied by an independent complex normal random variable. This convergence has been extended to the $L^1$ phase by Najnudel, Paquette, Simm and Vu. In the present article, we prove that this convergence further extends to the critical case $\beta = 2$, which corresponds to the limiting coefficients of the characteristic polynomial of the Circular Unitary Ensemble. We also prove the joint convergence of consecutive Fourier coefficients, and we derive convergence in distribution of the secular coefficients of the Circular Unitary Ensemble with index growing sufficiently slowly with the dimension.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 14:10:53 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Atherfold",
"Christopher",
""
],
[
"Najnudel",
"Joseph",
""
]
] |
2508.13883
|
Ilya Vilkoviskiy
|
Ilya Vilkoviskiy, Kirill Matirko
|
Properties of the temporal transfer matrix in integrable Floquet circuits
| null | null | null | null |
math-ph math.MP quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
One possible approach to studying non-equilibrium dynamics is the so-called influence matrix (IM) formalism. The influence matrix can be viewed as a quantum state that encodes complete information about the non-equilibrium dynamics of a boundary degree of freedom. It has been shown that the IM is the unique stationary point of the temporal transfer matrix. This transfer matrix, however, is non-diagonalizable and exhibits a non-trivial Jordan block structure. In this article, we demonstrate that, in the case of an integrable XXZ spin chain, the temporal transfer matrix itself is integrable and can be embedded into a family of commuting operators. We further provide the exact expression for the IM as a particular limit of a Bethe wavefunction, with the corresponding Bethe roots given explicitly. We also focus on the special case of the free-fermionic XX chain. In this setting, we uncover additional local integrals of motion, which enable us to analyze the dimensions and structure of the Jordan blocks, as well as the locality properties of the IM. Moreover, we construct a basis of quasi-local creation operators that generate the IM from the vacuum state.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 14:44:41 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 14:46:29 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Vilkoviskiy",
"Ilya",
""
],
[
"Matirko",
"Kirill",
""
]
] |
2508.13927
|
Anthony Lopes Temporao
|
Anthony Lopes Temporao, Mickael Tempor\~ao, Corentin Vande Kerckhove, Flavio Abreu Araujo
|
Towards a general diffusion-based information quality assessment model
|
24 pages, 3 figures
| null | null | null |
cs.SI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The rapid and unregulated dissemination of information in the digital era has amplified the global "infodemic," complicating the identification of high quality information. We present a lightweight, interpretable and non-invasive framework for assessing information quality based solely on diffusion dynamics, demonstrated here in the context of academic publications. Using a heterogeneous dataset of 29,264 sciences, technology, engineering, mathematics (STEM) and social science papers from ArnetMiner and OpenAlex, we model the diffusion network of each paper as a set of three theoretically motivated features: diversity, timeliness, and salience. A Generalized Additive Model (GAM) trained on these features achieved Pearson correlations of 0.8468 for next-year citation gain and up to 97.8% accuracy in predicting high-impact papers. Feature relevance studies reveal timeliness and salience as the most robust predictors, while diversity offers less stable benefits in the academic setting but may be more informative in social media contexts. The framework's transparency, domain-agnostic design, and minimal feature requirements position it as a scalable tool for global information quality assessment, opening new avenues for moving beyond binary credibility labels toward richer, diffusion-informed evaluation metrics.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 15:21:37 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 08:09:52 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Temporao",
"Anthony Lopes",
""
],
[
"Temporão",
"Mickael",
""
],
[
"Kerckhove",
"Corentin Vande",
""
],
[
"Araujo",
"Flavio Abreu",
""
]
] |
2508.13954
|
Samura\'i Gomes de Aguiar Brito
|
J. V. S. Scursulim, Gabriel Mattos Langeloh, Victor Leme Beltran, Samura\'i Brito
|
Multiclass Portfolio Optimization via Variational Quantum Eigensolver with Dicke State Ansatz
| null | null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Combinatorial optimization is a fundamental challenge in various domains, with portfolio optimization standing out as a key application in finance. Despite numerous quantum algorithmic approaches proposed for this problem, most overlook a critical feature of realistic portfolios: diversification. In this work, we introduce a novel quantum framework for multiclass portfolio optimization that explicitly incorporates diversification by leveraging multiple parametrized Dicke states, simultaneously initialized to encode the diversification constraints , as an ansatz of the Variational Quantum Eigensolver. A key strength of this ansatz is that it initializes the quantum system in a superposition of only feasible states, inherently satisfying the constraints. This significantly reduces the search space and eliminates the need for penalty terms. In addition, we also analyze the impact of different classical optimizers in this hybrid quantum-classical approach. Our findings demonstrate that, when combined with the CMA-ES optimizer, the Dicke state ansatz achieves superior performance in terms of convergence rate, approximation ratio, and measurement probability. These results underscore the potential of this method to solve practical, diversification-aware portfolio optimization problems relevant to the financial sector.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 15:45:07 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 02:57:12 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Scursulim",
"J. V. S.",
""
],
[
"Langeloh",
"Gabriel Mattos",
""
],
[
"Beltran",
"Victor Leme",
""
],
[
"Brito",
"Samuraí",
""
]
] |
2508.13957
|
Andrea Atzori
|
Andrea Atzori, Fadi Boutros, Naser Damer
|
ViT-FIQA: Assessing Face Image Quality using Vision Transformers
|
Accepted at the IEEE/CVF International Conference on Computer Vision Workshops 2025 (ICCVW 2025)
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Face Image Quality Assessment (FIQA) aims to predict the utility of a face image for face recognition (FR) systems. State-of-the-art FIQA methods mainly rely on convolutional neural networks (CNNs), leaving the potential of Vision Transformer (ViT) architectures underexplored. This work proposes ViT-FIQA, a novel approach that extends standard ViT backbones, originally optimized for FR, through a learnable quality token designed to predict a scalar utility score for any given face image. The learnable quality token is concatenated with the standard image patch tokens, and the whole sequence is processed via global self-attention by the ViT encoders to aggregate contextual information across all patches. At the output of the backbone, ViT-FIQA branches into two heads: (1) the patch tokens are passed through a fully connected layer to learn discriminative face representations via a margin-penalty softmax loss, and (2) the quality token is fed into a regression head to learn to predict the face sample's utility. Extensive experiments on challenging benchmarks and several FR models, including both CNN- and ViT-based architectures, demonstrate that ViT-FIQA consistently achieves top-tier performance. These results underscore the effectiveness of transformer-based architectures in modeling face image utility and highlight the potential of ViTs as a scalable foundation for future FIQA research https://cutt.ly/irHlzXUC.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 15:50:07 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 00:59:44 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Atzori",
"Andrea",
""
],
[
"Boutros",
"Fadi",
""
],
[
"Damer",
"Naser",
""
]
] |
2508.13968
|
Elias Stengel-Eskin
|
Tianyi Niu, Jaemin Cho, Elias Stengel-Eskin, Mohit Bansal
|
RotBench: Evaluating Multimodal Large Language Models on Identifying Image Rotation
|
20 pages. Code and data: https://github.com/tianyiniu/RotBench
| null | null | null |
cs.CV cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate to what extent Multimodal Large Language Models (MLLMs) can accurately identify the orientation of input images rotated 0{\deg}, 90{\deg}, 180{\deg}, and 270{\deg}. This task demands robust visual reasoning capabilities to detect rotational cues and contextualize spatial relationships within images, regardless of their orientation. To evaluate MLLMs on these abilities, we introduce RotBench -- a 350-image manually-filtered benchmark comprising lifestyle, portrait, and landscape images. Despite the relatively simple nature of this task, we show that several state-of-the-art open and proprietary MLLMs, including GPT-5, o3, and Gemini-2.5-Pro, do not reliably identify rotation in input images. Providing models with auxiliary information -- including captions, depth maps, and more -- or using chain-of-thought prompting offers only small and inconsistent improvements. Our results indicate that most models are able to reliably identify right-side-up (0{\deg}) images, while certain models are able to identify upside-down (180{\deg}) images. None can reliably distinguish between 90{\deg} and 270{\deg}. Simultaneously showing the image rotated in different orientations leads to moderate performance gains for reasoning models, while a modified setup using voting improves the performance of weaker models. We further show that fine-tuning does not improve models' ability to distinguish 90{\deg} and 270{\deg} rotations, despite substantially improving the identification of 180{\deg} images. Together, these results reveal a significant gap between MLLMs' spatial reasoning capabilities and human perception in identifying rotation.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 15:58:25 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 17:53:09 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Niu",
"Tianyi",
""
],
[
"Cho",
"Jaemin",
""
],
[
"Stengel-Eskin",
"Elias",
""
],
[
"Bansal",
"Mohit",
""
]
] |
2508.14029
|
Xiao Liang
|
Xiao Liang, Zhongzhi Li, Yeyun Gong, Yelong Shen, Ying Nian Wu, Zhijiang Guo, Weizhu Chen
|
Beyond Pass@1: Self-Play with Variational Problem Synthesis Sustains RLVR
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as a key paradigm for post-training Large Language Models (LLMs), particularly for complex reasoning tasks. However, vanilla RLVR training has been shown to improve Pass@1 performance at the expense of policy entropy, leading to reduced generation diversity and limiting the Pass@k performance, which typically represents the upper bound of LLM reasoning capability. In this paper, we systematically analyze the policy's generation diversity from the perspective of training problems and find that augmenting and updating training problems helps mitigate entropy collapse during training. Based on these observations, we propose an online Self-play with Variational problem Synthesis (SvS) strategy for RLVR training, which uses the policy's correct solutions to synthesize variational problems while ensuring their reference answers remain identical to the originals. This self-improving strategy effectively maintains policy entropy during training and substantially improves Pass@k compared with standard RLVR, sustaining prolonged improvements and achieving absolute gains of 18.3% and 22.8% in Pass@32 performance on the competition-level AIME24 and AIME25 benchmarks. Experiments on 12 reasoning benchmarks across varying model sizes from 3B to 32B consistently demonstrate the generalizability and robustness of SvS.
|
[
{
"version": "v1",
"created": "Tue, 19 Aug 2025 17:42:45 GMT"
},
{
"version": "v2",
"created": "Wed, 20 Aug 2025 01:21:25 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Liang",
"Xiao",
""
],
[
"Li",
"Zhongzhi",
""
],
[
"Gong",
"Yeyun",
""
],
[
"Shen",
"Yelong",
""
],
[
"Wu",
"Ying Nian",
""
],
[
"Guo",
"Zhijiang",
""
],
[
"Chen",
"Weizhu",
""
]
] |
2508.14043
|
Rahul Shukla
|
Digvijay Singh, Rahul Shukla, Karunesh Kumar Singh
|
A comparative study of some wavelet and sampling operators on various features of an image
|
15 pages
| null | null | null |
cs.CV math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This research includes the study of some positive sampling Kantorovich operators (SK operators) and their convergence properties. A comprehensive analysis of both local and global approximation properties is presented using sampling Kantorovich (SK), Gaussian, Bilateral and the thresholding wavelet-based operators in the framework of SK-operators. Explicitly, we start the article by introducing the basic terminology and state the fundamental theorem of approximation (FTA) by imposing the various required conditions corresponding to the various defined operators. We measure the error and study the other mathematical parameters such as the mean square error (MSE), the speckle index (SI), the speckle suppression index (SSI), the speckle mean preservation index (SMPI), and the equivalent number of looks (ENL) at various levels of resolution parameters. The nature of these operators are demonstrated via an example under ideal conditions in tabulated form at a certain level of samples. Eventually, another numerical example is illustrated to discuss the region of interest (ROI) via SI, SSI and SMPI of 2D Shepp-Logan Phantom taken slice from the 3D image, which gives the justification of the fundamental theorem of approximation (FTA). At the end of the derivation and illustrations we observe that the various operators have their own significance while studying the various features of the image because of the uneven nature of an image (non-ideal condition). Therefore, to some extent, some operators work well and some do not for some specific features of the image.
|
[
{
"version": "v1",
"created": "Thu, 03 Jul 2025 18:43:17 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Singh",
"Digvijay",
""
],
[
"Shukla",
"Rahul",
""
],
[
"Singh",
"Karunesh Kumar",
""
]
] |
2508.14044
|
Tao Zhang
|
Ying-Hua Lu, Tao Zhang
|
Cohomology theory for matched pairs of Leibniz algebras and its applications
|
28pages
| null | null | null |
math.RA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This paper introduces the concept of representations for matched pairs of Leibniz algebras and develops corresponding cohomology theories, examining how cohomology groups can be applied in the context of matched pairs of Leibniz algebras. Initially, we utilize the cohomology theory to investigate infinitesimal deformations and abelian extensions of matched pairs of Leibniz algebras. Next, given an abelian extension of a matched pair of Leibniz algebras and its representation, we explore the inducibility of pairs of automorphisms, deriving both necessary and sufficient conditions for the inducibility problem. Lastly, we delve deeper into the inducibility of pairs of automorphisms using the Wells exact sequences, offering a clear framework for studying the inducibility of pairs of automorphisms.
|
[
{
"version": "v1",
"created": "Sun, 20 Jul 2025 03:33:18 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Lu",
"Ying-Hua",
""
],
[
"Zhang",
"Tao",
""
]
] |
2508.14045
|
Rafail Admitos Passadakis
|
Admitos Passadakis, Yingjin Song and Albert Gatt
|
From Image Captioning to Visual Storytelling
|
16 pages (including references), 5 figures and 6 tables
| null | null | null |
cs.CL cs.CV
|
http://creativecommons.org/licenses/by-sa/4.0/
|
Visual Storytelling is a challenging multimodal task between Vision & Language, where the purpose is to generate a story for a stream of images. Its difficulty lies on the fact that the story should be both grounded to the image sequence but also narrative and coherent. The aim of this work is to balance between these aspects, by treating Visual Storytelling as a superset of Image Captioning, an approach quite different compared to most of prior relevant studies. This means that we firstly employ a vision-to-language model for obtaining captions of the input images, and then, these captions are transformed into coherent narratives using language-to-language methods. Our multifarious evaluation shows that integrating captioning and storytelling under a unified framework, has a positive impact on the quality of the produced stories. In addition, compared to numerous previous studies, this approach accelerates training time and makes our framework readily reusable and reproducible by anyone interested. Lastly, we propose a new metric/tool, named ideality, that can be used to simulate how far some results are from an oracle model, and we apply it to emulate human-likeness in visual storytelling.
|
[
{
"version": "v1",
"created": "Thu, 31 Jul 2025 16:44:23 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Passadakis",
"Admitos",
""
],
[
"Song",
"Yingjin",
""
],
[
"Gatt",
"Albert",
""
]
] |
2508.14046
|
Araf Rahman
|
Araf Rahman (1), M. Sabbir Salek (1), Mashrur Chowdhury (1) and Wayne A. Sarasua (1) ((1) Clemson University)
|
A Digital Twin-Based Simulation Framework for Safe Curve Speed Estimation Using Unity
|
20 pages, 7 figures, 2 tables Paper submitted for presentation at the Transportation Research Board 105th Annual Meeting and publication in Transportation Research Record. Under review for both cases
| null | null | null |
eess.SY cs.SY
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Horizontal curves are often associated with roadway crashes due to speed misjudgment and loss of control. With the growing adoption of autonomous and connected vehicles, the accurate estimation of safe speed at curves is becoming increasingly important. The widely used AASHTO design method for safe curve speed estimation relies on an analytical equation based on a simplified point mass model, which often uses conservative parameters to account for vehicular and environmental variations. This paper presents a digital twin-based framework for estimating safe speed at curves using a physics-driven virtual environment developed in the Unity engine. In this framework, a real-world horizontal road curve is selected, and vehicle speed data are collected using a radar gun under various weather conditions. A 3D model of the road curve is constructed in a Unity environment using roadway geometric and elevation data. A parameterized vehicle model is integrated, allowing for variations in mass, acceleration, and center of gravity to reflect different vehicle types and loading scenarios. This simulation identifies the maximum safe speed at which a vehicle can traverse the given curve, providing a more vehicle and environment-specific estimate of the safe operating speed. The study validated that the safe curve speed estimates generated by the simulation were consistent with the real-world speed values observed at a curve. This study demonstrates how a physics-based digital twin can estimate a safer and more adaptive operating speed for vehicles traversing horizontal curves.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 16:38:15 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Rahman",
"Araf",
"",
"Clemson University"
],
[
"Salek",
"M. Sabbir",
"",
"Clemson University"
],
[
"Chowdhury",
"Mashrur",
"",
"Clemson University"
],
[
"Sarasua",
"Wayne A.",
"",
"Clemson University"
]
] |
2508.14048
|
Sheng Li Dr.
|
Pengcheng Wang, Sheng Li, Takahiro Shinozaki
|
RAG-Boost: Retrieval-Augmented Generation Enhanced LLM-based Speech Recognition
|
accepted at Interspeech2025 MLC-SLM Challenge workshop (task I system description)
| null | null | null |
eess.AS cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we propose RAG-Boost (ST-ShinozakiLab Task I system), which enhances the baseline LLM-based ASR system of the MLC-SLM Challenge (task I) with a retrieval-augmented generation (RAG) module on the fly. Each partial ASR hypothesis queries a vector store of audio-text pairs and domain terms, and the retrieved results are fused with the live ASR hypotheses to fix recognition errors. The fused hypotheses are passed to the LLM, yielding improved responses.
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 17:53:11 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wang",
"Pengcheng",
""
],
[
"Li",
"Sheng",
""
],
[
"Shinozaki",
"Takahiro",
""
]
] |
2508.14049
|
Raghav Prabhakar
|
Jaskaran Singh, Amartya Roy Chowdhury, Raghav Prabhakar and Varshul C. W
|
MahaTTS: A Unified Framework for Multilingual Text-to-Speech Synthesis
| null | null | null | null |
eess.AS cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Current Text-to-Speech models pose a multilingual challenge, where most of the models traditionally focus on English and European languages, thereby hurting the potential to provide access to information to many more people. To address this gap, we introduce MahaTTS-v2 a Multilingual Multi-speaker Text-To-Speech (TTS) system that has excellent multilingual expressive capabilities in Indic languages. The model has been trained on around 20K hours of data specifically focused on Indian languages. Our approach leverages Wav2Vec2.0 tokens for semantic extraction, and a Language Model (LM) for text-to-semantic modeling. Additionally, we have used a Conditional Flow Model (CFM) for semantics to melspectogram generation. The experimental results indicate the effectiveness of the proposed approach over other frameworks. Our code is available at https://github.com/dubverse-ai/MahaTTSv2
|
[
{
"version": "v1",
"created": "Tue, 05 Aug 2025 20:49:04 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Singh",
"Jaskaran",
""
],
[
"Chowdhury",
"Amartya Roy",
""
],
[
"Prabhakar",
"Raghav",
""
],
[
"W",
"Varshul C.",
""
]
] |
2508.14050
|
Jennifer Eigenbrode
|
Jennifer L. Eigenbrode and Luoth Chou
|
The Importance of Aqueous Metabolites in the Martian Subsurface for Understanding Habitability, Organic Chemical Evolution, and Potential Biology
|
a white paper for MEPAG SFL-SAG
| null | null | null |
astro-ph.IM q-bio.BM
|
http://creativecommons.org/licenses/by/4.0/
|
Aqueous metabolites in terrestrial subsurface environments provide critical analog frameworks for assessing the habitability of Martian subsurface ice. On Earth, they play critical roles in sustaining microbial life within soils, permafrost, and groundwater environments and their availability shape microbial community compositions, activity, and adaptability to changes in environmental conditions, enabling communities to persist over millennial timescales. The counterpart to aqueous-soluble organics is the insoluble organic matter pool that makes up the largest portion of organic matter in natural samples and includes most types of organic signatures indicative of biological processes. Employing a range of sample preparation, molecular separation, detection, and imaging techniques enables the characterization of both labile (i.e., soluble and reactive) and recalcitrant (i.e., insoluble, non-reactive; include macromolecules) organic pools. Multiple orthogonal analytical modalities strengthen interpretations of signatures that we associate with biology as we know it and don't know it, by constraining possible abiotic sources, validating measurements across distinct techniques, and ensuring flexibility to interrogate diverse organic chemistries encountered in Martian subsurface environments. This holistic triage approach aligns with the priorities articulated in the Mars Exploration Program Analysis Group's Search for Life -Science Analysis Group (SFL-SAG) Charter for a medium-class Mars mission focused on extant life detection.
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 13:06:04 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Eigenbrode",
"Jennifer L.",
""
],
[
"Chou",
"Luoth",
""
]
] |
2508.14051
|
John Lalor
|
Kezia Oketch, John P. Lalor, Ahmed Abbasi
|
Benchmarking Sociolinguistic Diversity in Swahili NLP: A Taxonomy-Guided Approach
| null | null | null | null |
cs.CL cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
We introduce the first taxonomy-guided evaluation of Swahili NLP, addressing gaps in sociolinguistic diversity. Drawing on health-related psychometric tasks, we collect a dataset of 2,170 free-text responses from Kenyan speakers. The data exhibits tribal influences, urban vernacular, code-mixing, and loanwords. We develop a structured taxonomy and use it as a lens for examining model prediction errors across pre-trained and instruction-tuned language models. Our findings advance culturally grounded evaluation frameworks and highlight the role of sociolinguistic variation in shaping model performance.
|
[
{
"version": "v1",
"created": "Wed, 06 Aug 2025 20:10:11 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Oketch",
"Kezia",
""
],
[
"Lalor",
"John P.",
""
],
[
"Abbasi",
"Ahmed",
""
]
] |
2508.14053
|
Jinwei Tang
|
Jinwei Tang, Jiayin Qin, Nuo Xu, Pragnya Sudershan Nalla, Yu Cao, Yang (Katie) Zhao, Caiwen Ding
|
MAHL: Multi-Agent LLM-Guided Hierarchical Chiplet Design with Adaptive Debugging
| null | null | null | null |
cs.AR cs.AI cs.MA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As program workloads (e.g., AI) increase in size and algorithmic complexity, the primary challenge lies in their high dimensionality, encompassing computing cores, array sizes, and memory hierarchies. To overcome these obstacles, innovative approaches are required. Agile chip design has already benefited from machine learning integration at various stages, including logic synthesis, placement, and routing. With Large Language Models (LLMs) recently demonstrating impressive proficiency in Hardware Description Language (HDL) generation, it is promising to extend their abilities to 2.5D integration, an advanced technique that saves area overhead and development costs. However, LLM-driven chiplet design faces challenges such as flatten design, high validation cost and imprecise parameter optimization, which limit its chiplet design capability. To address this, we propose MAHL, a hierarchical LLM-based chiplet design generation framework that features six agents which collaboratively enable AI algorithm-hardware mapping, including hierarchical description generation, retrieval-augmented code generation, diverseflow-based validation, and multi-granularity design space exploration. These components together enhance the efficient generation of chiplet design with optimized Power, Performance and Area (PPA). Experiments show that MAHL not only significantly improves the generation accuracy of simple RTL design, but also increases the generation accuracy of real-world chiplet design, evaluated by Pass@5, from 0 to 0.72 compared to conventional LLMs under the best-case scenario. Compared to state-of-the-art CLARIE (expert-based), MAHL achieves comparable or even superior PPA results under certain optimization objectives.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 05:47:31 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tang",
"Jinwei",
"",
"Katie"
],
[
"Qin",
"Jiayin",
"",
"Katie"
],
[
"Xu",
"Nuo",
"",
"Katie"
],
[
"Nalla",
"Pragnya Sudershan",
"",
"Katie"
],
[
"Cao",
"Yu",
"",
"Katie"
],
[
"Yang",
"",
"",
"Katie"
],
[
"Zhao",
"",
""
],
[
"Ding",
"Caiwen",
""
]
] |
2508.14054
|
Yiran Rex Ma
|
Yiran Rex Ma
|
Contrastive Analysis of Constituent Order Preferences Within Adverbial Roles in English and Chinese News: A Large-Language-Model-Driven Approach
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Based on comparable English-Chinese news corpora annotated by Large Language Model (LLM), this paper attempts to explore the differences in constituent order of English-Chinese news from the perspective of functional chunks with adverbial roles, and analyze their typical positional preferences and distribution patterns. It is found that: (1) English news prefers linear narrative of core information first, and functional chunks are mostly post-positioned, while Chinese news prefers overall presentation mode of background first, and functional chunks are often pre-positioned; (2) In SVO structure, both English and Chinese news show differences in the distribution of functional chunks, but the tendency of Chinese pre-positioning is more significant, while that of English post-positioning is relatively mild; (3) When function blocks are co-occurring, both English and Chinese news show high flexibility, and the order adjustment is driven by information and pragmatic purposes. The study reveals that word order has both systematic preference and dynamic adaptability, providing new empirical support for contrastive study of English-Chinese information structure.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 11:20:22 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ma",
"Yiran Rex",
""
]
] |
2508.14055
|
Tim Luka Horstmann
|
Tim Luka Horstmann, Baptiste Geisenberger, Mehwish Alam
|
T-REX: Table -- Refute or Entail eXplainer
| null | null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Verifying textual claims against structured tabular data is a critical yet challenging task in Natural Language Processing with broad real-world impact. While recent advances in Large Language Models (LLMs) have enabled significant progress in table fact-checking, current solutions remain inaccessible to non-experts. We introduce T-REX (T-REX: Table -- Refute or Entail eXplainer), the first live, interactive tool for claim verification over multimodal, multilingual tables using state-of-the-art instruction-tuned reasoning LLMs. Designed for accuracy and transparency, T-REX empowers non-experts by providing access to advanced fact-checking technology. The system is openly available online.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 19:00:36 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Horstmann",
"Tim Luka",
""
],
[
"Geisenberger",
"Baptiste",
""
],
[
"Alam",
"Mehwish",
""
]
] |
2508.14056
|
Sepideh Entezari Maleki
|
Sepideh Entezari Maleki, Mohammadreza Pourreza, Davood Rafiei
|
Confidence Estimation for Text-to-SQL in Large Language Models
| null | null | null | null |
cs.CL cs.DB
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Confidence estimation for text-to-SQL aims to assess the reliability of model-generated SQL queries without having access to gold answers. We study this problem in the context of large language models (LLMs), where access to model weights and gradients is often constrained. We explore both black-box and white-box confidence estimation strategies, evaluating their effectiveness on cross-domain text-to-SQL benchmarks. Our evaluation highlights the superior performance of consistency-based methods among black-box models and the advantage of SQL-syntax-aware approaches for interpreting LLM logits in white-box settings. Furthermore, we show that execution-based grounding of queries provides a valuable supplementary signal, improving the effectiveness of both approaches.
|
[
{
"version": "v1",
"created": "Fri, 08 Aug 2025 23:09:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Maleki",
"Sepideh Entezari",
""
],
[
"Pourreza",
"Mohammadreza",
""
],
[
"Rafiei",
"Davood",
""
]
] |
2508.14057
|
Gladston Moreira
|
Pablo G. Almeida, Guilherme A. L. Silva, Val\'eria Santos, Gladston Moreira, Pedro Silva and Eduardo Luz
|
Deep Learning for School Dropout Detection: A Comparison of Tabular and Graph-Based Models for Predicting At-Risk Students
|
12 pages
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Student dropout is a significant challenge in educational systems worldwide, leading to substantial social and economic costs. Predicting students at risk of dropout allows for timely interventions. While traditional Machine Learning (ML) models operating on tabular data have shown promise, Graph Neural Networks (GNNs) offer a potential advantage by capturing complex relationships inherent in student data if structured as graphs. This paper investigates whether transforming tabular student data into graph structures, primarily using clustering techniques, enhances dropout prediction accuracy. We compare the performance of GNNs (a custom Graph Convolutional Network (GCN) and GraphSAGE) on these generated graphs against established tabular models (Random Forest (RF), XGBoost, and TabNet) using a real-world student dataset. Our experiments explore various graph construction strategies based on different clustering algorithms (K-Means, HDBSCAN) and dimensionality reduction techniques (Principal Component Analysis (PCA), Uniform Manifold Approximation and Projection (UMAP)). Our findings demonstrate that a specific GNN configuration, GraphSAGE on a graph derived from PCA-KMeans clustering, achieved superior performance, notably improving the macro F1-score by approximately 7 percentage points and accuracy by nearly 2 percentage points over the strongest tabular baseline (XGBoost). However, other GNN configurations and graph construction methods did not consistently surpass tabular models, emphasizing the critical role of the graph generation strategy and GNN architecture selection. This highlights both the potential of GNNs and the challenges in optimally transforming tabular data for graph-based learning in this domain.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 01:19:32 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Almeida",
"Pablo G.",
""
],
[
"Silva",
"Guilherme A. L.",
""
],
[
"Santos",
"Valéria",
""
],
[
"Moreira",
"Gladston",
""
],
[
"Silva",
"Pedro",
""
],
[
"Luz",
"Eduardo",
""
]
] |
2508.14058
|
Jingmao Zhang
|
Jingmao Zhang, Zhiting Zhao, Yunqi Lin, Jianghong Ma, Tianjun Wei, Haijun Zhang, Xiaofeng Zhang
|
Dual-Phase Playtime-guided Recommendation: Interest Intensity Exploration and Multimodal Random Walks
|
Accepted for publication at ACM Multimedia (ACM MM) 2025. 10 pages, 5 figures. Code and dataset: https://github.com/zqxwcevrtyui/DP2Rec
| null | null | null |
cs.IR cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The explosive growth of the video game industry has created an urgent need for recommendation systems that can scale with expanding catalogs and maintain user engagement. While prior work has explored accuracy and diversity in recommendations, existing models underutilize playtime, a rich behavioral signal unique to gaming platforms, and overlook the potential of multimodal information to enhance diversity. In this paper, we propose DP2Rec, a novel Dual-Phase Playtime-guided Recommendation model designed to jointly optimize accuracy and diversity. First, we introduce a playtime-guided interest intensity exploration module that separates strong and weak preferences via dual-beta modeling, enabling fine-grained user profiling and more accurate recommendations. Second, we present a playtime-guided multimodal random walks module that simulates player exploration using transitions guided by both playtime-derived interest similarity and multimodal semantic similarity. This mechanism preserves core preferences while promoting cross-category discovery through latent semantic associations and adaptive category balancing. Extensive experiments on a real-world game dataset show that DP2Rec outperforms existing methods in both recommendation accuracy and diversity.
|
[
{
"version": "v1",
"created": "Sat, 09 Aug 2025 13:51:00 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhang",
"Jingmao",
""
],
[
"Zhao",
"Zhiting",
""
],
[
"Lin",
"Yunqi",
""
],
[
"Ma",
"Jianghong",
""
],
[
"Wei",
"Tianjun",
""
],
[
"Zhang",
"Haijun",
""
],
[
"Zhang",
"Xiaofeng",
""
]
] |
2508.14059
|
Frank Yang
|
Mengyang Cao, Frank F. Yang, Yi Jin, Yijun Yan
|
Graph Neural Network for Product Recommendation on the Amazon Co-purchase Graph
|
15 pages, 5 figures, preprint
| null | null | null |
cs.IR cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Identifying relevant information among massive volumes of data is a challenge for modern recommendation systems. Graph Neural Networks (GNNs) have demonstrated significant potential by utilizing structural and semantic relationships through graph-based learning. This study assessed the abilities of four GNN architectures, LightGCN, GraphSAGE, GAT, and PinSAGE, on the Amazon Product Co-purchase Network under link prediction settings. We examined practical trade-offs between architectures, model performance, scalability, training complexity and generalization. The outcomes demonstrated each model's performance characteristics for deploying GNN in real-world recommendation scenarios.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 02:12:04 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Cao",
"Mengyang",
""
],
[
"Yang",
"Frank F.",
""
],
[
"Jin",
"Yi",
""
],
[
"Yan",
"Yijun",
""
]
] |
2508.14060
|
Kartik Pandey
|
Kartik Pandey, Arun Balasubramanian, Debasis Samanta
|
Activity Coefficient-based Channel Selection for Electroencephalogram: A Task-Independent Approach
| null | null | null | null |
q-bio.NC cs.CV cs.HC cs.LG eess.SP
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Electroencephalogram (EEG) signals have gained widespread adoption in brain-computer interface (BCI) applications due to their non-invasive, low-cost, and relatively simple acquisition process. The demand for higher spatial resolution, particularly in clinical settings, has led to the development of high-density electrode arrays. However, increasing the number of channels introduces challenges such as cross-channel interference and computational overhead. To address these issues, modern BCI systems often employ channel selection algorithms. Existing methods, however, are typically task-specific and require re-optimization for each new application. This work proposes a task-agnostic channel selection method, Activity Coefficient-based Channel Selection (ACCS), which uses a novel metric called the Channel Activity Coefficient (CAC) to quantify channel utility based on activity levels. By selecting the top 16 channels ranked by CAC, ACCS achieves up to 34.97% improvement in multi-class classification accuracy. Unlike traditional approaches, ACCS identifies a reusable set of informative channels independent of the downstream task or model, making it highly adaptable for diverse EEG-based applications.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 04:40:41 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Pandey",
"Kartik",
""
],
[
"Balasubramanian",
"Arun",
""
],
[
"Samanta",
"Debasis",
""
]
] |
2508.14061
|
Anurag Kumar Ojha
|
Anurag Kumar Ojha
|
GPT-2 as a Compression Preprocessor: Improving Gzip for Structured Text Domains
| null | null | null | null |
cs.IR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In the modern era, large volumes of data are being produced continuously, especially in domain-specific fields such as medical records and clinical files, defence logs and HTML-based web traffic. Data with such volume and complexity needs to be compressed before storing and transmitting efficiently. Data compression has gained significant attention from modern researchers, resulting in the development of fast and efficient compression algorithms such as Gzip. However, since gzip works on the principle of repetition of binary patterns, one of the limitations of gzip is that domain-specific formats like JSON, XML, HTML, and log files, while structured, may have semantic repetition but not syntactic repetition, which gzip finds difficult to compress. In this article, we propose a GPT-based preprocessor for such domain-specific files. We propose a pipeline made up of GPT-2 taking domain-specific files as input, which pattern-based compressors like gzip find difficult to work on. The preprocessor results are output in a file that is designed for compressors like gzip. After preprocessing, the gzip works on the other end of the pipeline and compresses the data as usual. We used different types of both real-world and synthetically generated data, such as logs and HTML files, for the experiment of the proposed model. We found promising results and an improvement of the Defence logs by 0.34 per cent and HTML files by 5.8 per cent.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 04:53:22 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ojha",
"Anurag Kumar",
""
]
] |
2508.14062
|
Akshaya Balaji
|
Badrinath Ramakrishnan, Akshaya Balaji
|
Assessing and Mitigating Data Memorization Risks in Fine-Tuned Large Language Models
|
14 pages, 2 figures. Code and experimental framework available at https://github.com/akshayaaa10/llm-privacy-research
| null | null | null |
cs.CL cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks, but their tendency to memorize training data poses significant privacy risks, particularly during fine-tuning processes. This paper presents a comprehensive empirical analysis of data memorization in fine-tuned LLMs and introduces a novel multi-layered privacy protection framework. Through controlled experiments on modern LLM architectures including GPT-2, Phi-3, and Gemma-2, we demonstrate that fine-tuning with repeated sensitive data increases privacy leakage rates from baseline levels of 0-5% to 60-75%, representing a 64.2% average increase across tested models. We propose and rigorously evaluate four complementary privacy protection methods: semantic data deduplication, differential privacy during generation, entropy-based filtering, and pattern-based content filtering. Our experimental results show that these techniques can reduce data leakage to 0% while maintaining 94.7% of original model utility.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 10:26:55 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ramakrishnan",
"Badrinath",
""
],
[
"Balaji",
"Akshaya",
""
]
] |
2508.14063
|
Dvir Aran
|
Moran Sorka, Alon Gorenshtein, Dvir Aran and Shahar Shelly
|
A Multi-Agent Approach to Neurological Clinical Reasoning
| null | null | null | null |
cs.IR cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Large language models (LLMs) have shown promise in medical domains, but their ability to handle specialized neurological reasoning requires systematic evaluation. We developed a comprehensive benchmark using 305 questions from Israeli Board Certification Exams in Neurology, classified along three complexity dimensions: factual knowledge depth, clinical concept integration, and reasoning complexity. We evaluated ten LLMs using base models, retrieval-augmented generation (RAG), and a novel multi-agent system. Results showed significant performance variation. OpenAI-o1 achieved the highest base performance (90.9% accuracy), while specialized medical models performed poorly (52.9% for Meditron-70B). RAG provided modest benefits but limited effectiveness on complex reasoning questions. In contrast, our multi-agent framework, decomposing neurological reasoning into specialized cognitive functions including question analysis, knowledge retrieval, answer synthesis, and validation, achieved dramatic improvements, especially for mid-range models. The LLaMA 3.3-70B-based agentic system reached 89.2% accuracy versus 69.5% for its base model, with substantial gains on level 3 complexity questions. The multi-agent approach transformed inconsistent subspecialty performance into uniform excellence, addressing neurological reasoning challenges that persisted with RAG enhancement. We validated our approach using an independent dataset of 155 neurological cases from MedQA. Results confirm that structured multi-agent approaches designed to emulate specialized cognitive processes significantly enhance complex medical reasoning, offering promising directions for AI assistance in challenging clinical contexts.
|
[
{
"version": "v1",
"created": "Sun, 10 Aug 2025 14:52:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sorka",
"Moran",
""
],
[
"Gorenshtein",
"Alon",
""
],
[
"Aran",
"Dvir",
""
],
[
"Shelly",
"Shahar",
""
]
] |
2508.14064
|
Ziyang Ding
|
Yao Ding, Yuqing Wu, Ziyang Ding
|
An automatic patent literature retrieval system based on LLM-RAG
| null | null | null | null |
cs.IR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
With the acceleration of technological innovation efficient retrieval and classification of patent literature have become essential for intellectual property management and enterprise RD Traditional keyword and rulebased retrieval methods often fail to address complex query intents or capture semantic associations across technical domains resulting in incomplete and lowrelevance results This study presents an automated patent retrieval framework integrating Large Language Models LLMs with RetrievalAugmented Generation RAG technology The system comprises three components: 1) a preprocessing module for patent data standardization, 2) a highefficiency vector retrieval engine leveraging LLMgenerated embeddings, and 3) a RAGenhanced query module that combines external document retrieval with contextaware response generation Evaluations were conducted on the Google Patents dataset 20062024 containing millions of global patent records with metadata such as filing date domain and status The proposed gpt35turbo0125RAG configuration achieved 805 semantic matching accuracy and 92.1% recall surpassing baseline LLM methods by 28 percentage points The framework also demonstrated strong generalization in crossdomain classification and semantic clustering tasks These results validate the effectiveness of LLMRAG integration for intelligent patent retrieval providing a foundation for nextgeneration AIdriven intellectual property analysis platforms
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 02:39:16 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ding",
"Yao",
""
],
[
"Wu",
"Yuqing",
""
],
[
"Ding",
"Ziyang",
""
]
] |
2508.14065
|
Srilakshmi Madiraju
|
Madiraju Srilakshmi, Kartavya Kothari, Kamlesh Marathe, Vedavyas Chigurupati, Hitesh Kapoor
|
Personalized Contest Recommendation in Fantasy Sports
| null | null | null | null |
cs.IR cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In daily fantasy sports, players enter into "contests" where they compete against each other by building teams of athletes that score fantasy points based on what actually occurs in a real-life sports match. For any given sports match, there are a multitude of contests available to players, with substantial variation across 3 main dimensions: entry fee, number of spots, and the prize pool distribution. As player preferences are also quite heterogeneous, contest personalization is an important tool to match players with contests. This paper presents a scalable contest recommendation system, powered by a Wide and Deep Interaction Ranker (WiDIR) at its core. We productionized this system at our company, one of the large fantasy sports platforms with millions of daily contests and millions of players, where online experiments show a marked improvement over other candidate models in terms of recall and other critical business metrics.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 07:22:33 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Srilakshmi",
"Madiraju",
""
],
[
"Kothari",
"Kartavya",
""
],
[
"Marathe",
"Kamlesh",
""
],
[
"Chigurupati",
"Vedavyas",
""
],
[
"Kapoor",
"Hitesh",
""
]
] |
2508.14066
|
Lorenz Brehme
|
Lorenz Brehme, Benedikt Dornauer, Thomas Str\"ohle, Maximilian Ehrhart and Ruth Breu
|
Retrieval-Augmented Generation in Industry: An Interview Study on Use Cases, Requirements, Challenges, and Evaluation
|
This preprint was accepted for presentation at the 17th International Conference on Knowledge Discovery and Information Retrieval (KDIR25)
| null | null | null |
cs.IR cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Retrieval-Augmented Generation (RAG) is a well-established and rapidly evolving field within AI that enhances the outputs of large language models by integrating relevant information retrieved from external knowledge sources. While industry adoption of RAG is now beginning, there is a significant lack of research on its practical application in industrial contexts. To address this gap, we conducted a semistructured interview study with 13 industry practitioners to explore the current state of RAG adoption in real-world settings. Our study investigates how companies apply RAG in practice, providing (1) an overview of industry use cases, (2) a consolidated list of system requirements, (3) key challenges and lessons learned from practical experiences, and (4) an analysis of current industry evaluation methods. Our main findings show that current RAG applications are mostly limited to domain-specific QA tasks, with systems still in prototype stages; industry requirements focus primarily on data protection, security, and quality, while issues such as ethics, bias, and scalability receive less attention; data preprocessing remains a key challenge, and system evaluation is predominantly conducted by humans rather than automated methods.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 09:40:54 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Brehme",
"Lorenz",
""
],
[
"Dornauer",
"Benedikt",
""
],
[
"Ströhle",
"Thomas",
""
],
[
"Ehrhart",
"Maximilian",
""
],
[
"Breu",
"Ruth",
""
]
] |
2508.14067
|
Nandi Schoots
|
Sonakshi Chauhan, Maheep Chaudhary, Koby Choy, Samuel Nellessen, Nandi Schoots
|
Punctuation and Predicates in Language Models
| null | null | null | null |
cs.CL cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper we explore where information is collected and how it is propagated throughout layers in large language models (LLMs). We begin by examining the surprising computational importance of punctuation tokens which previous work has identified as attention sinks and memory aids. Using intervention-based techniques, we evaluate the necessity and sufficiency (for preserving model performance) of punctuation tokens across layers in GPT-2, DeepSeek, and Gemma. Our results show stark model-specific differences: for GPT-2, punctuation is both necessary and sufficient in multiple layers, while this holds far less in DeepSeek and not at all in Gemma. Extending beyond punctuation, we ask whether LLMs process different components of input (e.g., subjects, adjectives, punctuation, full sentences) by forming early static summaries reused across the network, or if the model remains sensitive to changes in these components across layers. Extending beyond punctuation, we investigate whether different reasoning rules are processed differently by LLMs. In particular, through interchange intervention and layer-swapping experiments, we find that conditional statements (if, then), and universal quantification (for all) are processed very differently. Our findings offer new insight into the internal mechanisms of punctuation usage and reasoning in LLMs and have implications for interpretability.
|
[
{
"version": "v1",
"created": "Mon, 11 Aug 2025 09:53:30 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chauhan",
"Sonakshi",
""
],
[
"Chaudhary",
"Maheep",
""
],
[
"Choy",
"Koby",
""
],
[
"Nellessen",
"Samuel",
""
],
[
"Schoots",
"Nandi",
""
]
] |
2508.14068
|
Chen Chen
|
Chen Chen, Jiaqi Yin, Cunxi Yu
|
Revisit Choice Network for Synthesis and Technology Mapping
|
Accepted by ICCAD 2025
| null | null | null |
cs.AR cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Choice network construction is a critical technique for alleviating structural bias issues in Boolean optimization, equivalence checking, and technology mapping. Previous works on lossless synthesis utilize independent optimization to generate multiple snapshots, and use simulation and SAT solvers to identify functionally equivalent nodes. These nodes are then merged into a subject graph with choice nodes. However, such methods often neglect the quality of these choices, raising the question of whether they truly contribute to effective technology mapping.
This paper introduces Cristal, a novel methodology and framework for constructing Boolean choice networks. Specifically, Cristal introduces a new flow of choice network-based synthesis and mapping, including representative logic cone search, structural mutation for generating diverse choice structures via equality saturation, and priority-ranking choice selection along with choice network construction and validation. Through these techniques, Cristal constructs fewer but higher-quality choices.
Our experimental results demonstrate that Cristal outperforms the state-of-the-art Boolean choice network construction implemented in ABC in the post-mapping stage, achieving average reductions of 3.85%/8.35% (area/delay) in delay-oriented mode, 0.11%/2.74% in area-oriented mode, and a 63.77% runtime reduction on large-scale cases across a diverse set of combinational circuits from the IWLS 2005, ISCAS'89, and EPFL benchmark suites.
|
[
{
"version": "v1",
"created": "Mon, 04 Aug 2025 10:09:10 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chen",
"Chen",
""
],
[
"Yin",
"Jiaqi",
""
],
[
"Yu",
"Cunxi",
""
]
] |
2508.14069
|
Hafiz Imtiaz
|
Chinmoy Biswas, Nafis Faisal, Vivek Chowdhury, Abrar Al-Shadid Abir, Sabir Mahmud, Mithon Rahman, Shaikh Anowarul Fattah, Hafiz Imtiaz
|
Load Forecasting on A Highly Sparse Electrical Load Dataset Using Gaussian Interpolation
|
Under review in Elsevier Electric Power Systems Research
| null | null | null |
cs.LG eess.SP
|
http://creativecommons.org/licenses/by/4.0/
|
Sparsity, defined as the presence of missing or zero values in a dataset, often poses a major challenge while operating on real-life datasets. Sparsity in features or target data of the training dataset can be handled using various interpolation methods, such as linear or polynomial interpolation, spline, moving average, or can be simply imputed. Interpolation methods usually perform well with Strict Sense Stationary (SSS) data. In this study, we show that an approximately 62\% sparse dataset with hourly load data of a power plant can be utilized for load forecasting assuming the data is Wide Sense Stationary (WSS), if augmented with Gaussian interpolation. More specifically, we perform statistical analysis on the data, and train multiple machine learning and deep learning models on the dataset. By comparing the performance of these models, we empirically demonstrate that Gaussian interpolation is a suitable option for dealing with load forecasting problems. Additionally, we demonstrate that Long Short-term Memory (LSTM)-based neural network model offers the best performance among a diverse set of classical and neural network-based models.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 03:15:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Biswas",
"Chinmoy",
""
],
[
"Faisal",
"Nafis",
""
],
[
"Chowdhury",
"Vivek",
""
],
[
"Abir",
"Abrar Al-Shadid",
""
],
[
"Mahmud",
"Sabir",
""
],
[
"Rahman",
"Mithon",
""
],
[
"Fattah",
"Shaikh Anowarul",
""
],
[
"Imtiaz",
"Hafiz",
""
]
] |
2508.14070
|
Ephraiem Sarabamoun
|
Ephraiem Sarabamoun
|
Special-Character Adversarial Attacks on Open-Source Language Model
| null | null | null | null |
cs.CR cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Large language models (LLMs) have achieved remarkable performance across diverse natural language processing tasks, yet their vulnerability to character-level adversarial manipulations presents significant security challenges for real-world deployments.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 03:42:59 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sarabamoun",
"Ephraiem",
""
]
] |
2508.14071
|
Bachtiar Herdianto
|
Bachtiar Herdianto, Romain Billot, Flavien Lucas, Marc Sevaux, and Daniele Vigo
|
Edge-Selector Model Applied for Local Search Neighborhood for Solving Vehicle Routing Problems
|
29 pages, 12 figures
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
This research proposes a hybrid Machine Learning and metaheuristic mechanism that is designed to solve Vehicle Routing Problems (VRPs). The main of our method is an edge solution selector model, which classifies solution edges to identify prohibited moves during the local search, hence guiding the search process within metaheuristic baselines. Two learning-based mechanisms are used to develop the edge selector: a simple tabular binary classifier and a Graph Neural Network (GNN). The tabular classifier employs Gradient Boosting Trees and Feedforward Neural Network as the baseline algorithms. Adjustments to the decision threshold are also applied to handle the class imbalance in the problem instance. An alternative mechanism employs the GNN to utilize graph structure for direct solution edge prediction, with the objective of guiding local search by predicting prohibited moves. These hybrid mechanisms are then applied in state-fo-the-art metaheuristic baselines. Our method demonstrates both scalability and generalizability, achieving performance improvements across different baseline metaheuristics, various problem sizes and variants, including the Capacitated Vehicle Routing Problem (CVRP) and CVRP with Time Windows (CVRPTW). Experimental evaluations on benchmark datasets up to 30,000 customer nodes, supported by pair-wise statistical analysis, verify the observed improvements.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 05:28:26 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Herdianto",
"Bachtiar",
""
],
[
"Billot",
"Romain",
""
],
[
"Lucas",
"Flavien",
""
],
[
"Sevaux",
"Marc",
""
],
[
"Vigo",
"Daniele",
""
]
] |
2508.14072
|
Anabel Yong
|
Anabel Yong
|
Multi-Objective Bayesian Optimization with Independent Tanimoto Kernel Gaussian Processes for Diverse Pareto Front Exploration
|
Masters of Science thesis
| null | null | null |
cs.LG physics.chem-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present GP-MOBO, a novel multi-objective Bayesian Optimization algorithm that advances the state-of-the-art in molecular optimization. Our approach integrates a fast minimal package for Exact Gaussian Processes (GPs) capable of efficiently handling the full dimensionality of sparse molecular fingerprints without the need for extensive computational resources. GP-MOBO consistently outperforms traditional methods like GP-BO by fully leveraging fingerprint dimensionality, leading to the identification of higher-quality and valid SMILES. Moreover, our model achieves a broader exploration of the chemical search space, as demonstrated by its superior proximity to the Pareto front in all tested scenarios. Empirical results from the DockSTRING dataset reveal that GP-MOBO yields higher geometric mean values across 20 Bayesian optimization iterations, underscoring its effectiveness and efficiency in addressing complex multi-objective optimization challenges with minimal computational overhead.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 06:27:36 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Yong",
"Anabel",
""
]
] |
2508.14074
|
Ruilin Zhang
|
Qian Zhang, Ruilin Zhang, Biaokai Zhu, Xun Han, Jun Xiao, Yifan Liu, Zhe Wang
|
GEPD:GAN-Enhanced Generalizable Model for EEG-Based Detection of Parkinson's Disease
|
Accepted by International Conference on Intelligent Computing(ICIC 2025)
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/publicdomain/zero/1.0/
|
Electroencephalography has been established as an effective method for detecting Parkinson's disease, typically diagnosed early.Current Parkinson's disease detection methods have shown significant success within individual datasets, however, the variability in detection methods across different EEG datasets and the small size of each dataset pose challenges for training a generalizable model for cross-dataset scenarios. To address these issues, this paper proposes a GAN-enhanced generalizable model, named GEPD, specifically for EEG-based cross-dataset classification of Parkinson's disease.First, we design a generative network that creates fusion EEG data by controlling the distribution similarity between generated data and real data.In addition, an EEG signal quality assessment model is designed to ensure the quality of generated data great.Second, we design a classification network that utilizes a combination of multiple convolutional neural networks to effectively capture the time-frequency characteristics of EEG signals, while maintaining a generalizable structure and ensuring easy convergence.This work is dedicated to utilizing intelligent methods to study pathological manifestations, aiming to facilitate the diagnosis and monitoring of neurological diseases.The evaluation results demonstrate that our model performs comparably to state-of-the-art models in cross-dataset settings, achieving an accuracy of 84.3% and an F1-score of 84.0%, showcasing the generalizability of the proposed model.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 08:37:14 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Zhang",
"Qian",
""
],
[
"Zhang",
"Ruilin",
""
],
[
"Zhu",
"Biaokai",
""
],
[
"Han",
"Xun",
""
],
[
"Xiao",
"Jun",
""
],
[
"Liu",
"Yifan",
""
],
[
"Wang",
"Zhe",
""
]
] |
2508.14075
|
Mieczys{\l}aw K{\l}opotek
|
Mieczys{\l}aw A. K{\l}opotek and S{\l}awomir T. Wierzcho\'n and Bart{\l}omiej Starosta and Piotr Borkowski and Dariusz Czerski and Eryk Laskowski
|
Explainable Graph Spectral Clustering For Text Embeddings
|
47 pages, 19 tables, 11 figures
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
In a previous paper, we proposed an introduction to the explainability of Graph Spectral Clustering results for textual documents, given that document similarity is computed as cosine similarity in term vector space.
In this paper, we generalize this idea by considering other embeddings of documents, in particular, based on the GloVe embedding idea.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 11:20:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kłopotek",
"Mieczysław A.",
""
],
[
"Wierzchoń",
"Sławomir T.",
""
],
[
"Starosta",
"Bartłomiej",
""
],
[
"Borkowski",
"Piotr",
""
],
[
"Czerski",
"Dariusz",
""
],
[
"Laskowski",
"Eryk",
""
]
] |
2508.14076
|
Mengdi Li
|
Mengdi Li, Guanqiao Chen, Xufeng Zhao, Haochen Wen, Shu Yang, Di Wang
|
PersRM-R1: Enhance Personalized Reward Modeling with Reinforcement Learning
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Reward models (RMs), which are central to existing post-training methods, aim to align LLM outputs with human values by providing feedback signals during fine-tuning. However, existing RMs struggle to capture nuanced, user-specific preferences, especially under limited data and across diverse domains. Thus, we introduce PersRM-R1, the first reasoning-based reward modeling framework specifically designed to identify and represent personal factors from only one or a few personal exemplars. To address challenges including limited data availability and the requirement for robust generalization, our approach combines synthetic data generation with a two-stage training pipeline consisting of supervised fine-tuning followed by reinforcement fine-tuning. Experimental results demonstrate that PersRM-R1 outperforms existing models of similar size and matches the performance of much larger models in both accuracy and generalizability, paving the way for more effective personalized LLMs.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 14:25:58 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Li",
"Mengdi",
""
],
[
"Chen",
"Guanqiao",
""
],
[
"Zhao",
"Xufeng",
""
],
[
"Wen",
"Haochen",
""
],
[
"Yang",
"Shu",
""
],
[
"Wang",
"Di",
""
]
] |
2508.14077
|
Sota Kudo
|
Sota Kudo
|
Label Smoothing is a Pragmatic Information Bottleneck
|
18 pages, 8 figures, published in Transactions on Machine Learning Research (TMLR), 2025
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This study revisits label smoothing via a form of information bottleneck. Under the assumption of sufficient model flexibility and no conflicting labels for the same input, we theoretically and experimentally demonstrate that the model output obtained through label smoothing explores the optimal solution of the information bottleneck. Based on this, label smoothing can be interpreted as a practical approach to the information bottleneck, enabling simple implementation. As an information bottleneck method, we experimentally show that label smoothing also exhibits the property of being insensitive to factors that do not contain information about the target, or to factors that provide no additional information about it when conditioned on another variable.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 14:50:23 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kudo",
"Sota",
""
]
] |
2508.14078
|
Mohamed Hassan Abdalla Idris
|
Mohamed Hassan Abdalla Idris, Jakub Marek Cebula, Jebraeel Gholinezhad, Shamsul Masum, Hongjie Ma
|
Out-of-Sample Hydrocarbon Production Forecasting: Time Series Machine Learning using Productivity Index-Driven Features and Inductive Conformal Prediction
| null | null | null | null |
cs.LG physics.data-an
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
This research introduces a new ML framework designed to enhance the robustness of out-of-sample hydrocarbon production forecasting, specifically addressing multivariate time series analysis. The proposed methodology integrates Productivity Index (PI)-driven feature selection, a concept derived from reservoir engineering, with Inductive Conformal Prediction (ICP) for rigorous uncertainty quantification. Utilizing historical data from the Volve (wells PF14, PF12) and Norne (well E1H) oil fields, this study investigates the efficacy of various predictive algorithms-namely Long Short-Term Memory (LSTM), Bidirectional LSTM (BiLSTM), Gated Recurrent Unit (GRU), and eXtreme Gradient Boosting (XGBoost) - in forecasting historical oil production rates (OPR_H). All the models achieved "out-of-sample" production forecasts for an upcoming future timeframe. Model performance was comprehensively evaluated using traditional error metrics (e.g., MAE) supplemented by Forecast Bias and Prediction Direction Accuracy (PDA) to assess bias and trend-capturing capabilities. The PI-based feature selection effectively reduced input dimensionality compared to conventional numerical simulation workflows. The uncertainty quantification was addressed using the ICP framework, a distribution-free approach that guarantees valid prediction intervals (e.g., 95% coverage) without reliance on distributional assumptions, offering a distinct advantage over traditional confidence intervals, particularly for complex, non-normal data. Results demonstrated the superior performance of the LSTM model, achieving the lowest MAE on test (19.468) and genuine out-of-sample forecast data (29.638) for well PF14, with subsequent validation on Norne well E1H. These findings highlight the significant potential of combining domain-specific knowledge with advanced ML techniques to improve the reliability of hydrocarbon production forecasts.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 19:14:46 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Idris",
"Mohamed Hassan Abdalla",
""
],
[
"Cebula",
"Jakub Marek",
""
],
[
"Gholinezhad",
"Jebraeel",
""
],
[
"Masum",
"Shamsul",
""
],
[
"Ma",
"Hongjie",
""
]
] |
2508.14079
|
Maxime Heuillet
|
Maxime Heuillet, Rishika Bhagwatkar, Jonas Ngnaw\'e, Yann Pequignot, Alexandre Larouche, Christian Gagn\'e, Irina Rish, Ola Ahmad, Audrey Durand
|
A Guide to Robust Generalization: The Impact of Architecture, Pre-training, and Optimization Strategy
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Deep learning models operating in the image domain are vulnerable to small input perturbations. For years, robustness to such perturbations was pursued by training models from scratch (i.e., with random initializations) using specialized loss objectives. Recently, robust fine-tuning has emerged as a more efficient alternative: instead of training from scratch, pretrained models are adapted to maximize predictive performance and robustness. To conduct robust fine-tuning, practitioners design an optimization strategy that includes the model update protocol (e.g., full or partial) and the specialized loss objective. Additional design choices include the architecture type and size, and the pretrained representation. These design choices affect robust generalization, which is the model's ability to maintain performance when exposed to new and unseen perturbations at test time. Understanding how these design choices influence generalization remains an open question with significant practical implications. In response, we present an empirical study spanning 6 datasets, 40 pretrained architectures, 2 specialized losses, and 3 adaptation protocols, yielding 1,440 training configurations and 7,200 robustness measurements across five perturbation types. To our knowledge, this is the most diverse and comprehensive benchmark of robust fine-tuning to date. While attention-based architectures and robust pretrained representations are increasingly popular, we find that convolutional neural networks pretrained in a supervised manner on large datasets often perform best. Our analysis both confirms and challenges prior design assumptions, highlighting promising research directions and offering practical guidance.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 19:38:35 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Heuillet",
"Maxime",
""
],
[
"Bhagwatkar",
"Rishika",
""
],
[
"Ngnawé",
"Jonas",
""
],
[
"Pequignot",
"Yann",
""
],
[
"Larouche",
"Alexandre",
""
],
[
"Gagné",
"Christian",
""
],
[
"Rish",
"Irina",
""
],
[
"Ahmad",
"Ola",
""
],
[
"Durand",
"Audrey",
""
]
] |
2508.14080
|
Guanghao Jin
|
Guanghao Jin, Jingpei Wu, Tianpei Guo, Yiyi Niu, Weidong Zhou, Guoyang Liu
|
KnowDR-REC: A Benchmark for Referring Expression Comprehension with Real-World Knowledge
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Referring Expression Comprehension (REC) is a popular multimodal task that aims to accurately detect target objects within a single image based on a given textual expression. However, due to the limitations of earlier models, traditional REC benchmarks either rely solely on intra-image cues or lack sufficiently fine-grained instance annotations, making them inadequate for evaluating the reasoning capabilities of Multi-modal Large Language Models (MLLMs). To address this gap, we propose a new benchmark, KnowDR-REC, characterized by three key features: Firstly, it is built upon real-world knowledge, requiring fine-grained multimodal reasoning across text and image. Secondly, the dataset includes elaborately constructed negative samples via fine-grained expression editing, designed to evaluate a model's robustness and anti-hallucination ability. Lastly, we introduce three novel evaluation metrics to systematically explore the model's internal reasoning process. We evaluate 16 state-of-the-art multimodal models on KnowDR-REC, with experimental results showing that existing MLLMs still struggle with knowledge-driven visual grounding tasks. Furthermore, we observe a decoupling between textual understanding and visual grounding in MLLMs, where many models are significantly influenced by memorized shortcut correlations, which severely affect their behavior on our benchmark and hinder genuine multimodal reasoning. We anticipate that the proposed benchmark will inspire future research towards developing more robust, interpretable, and knowledge-intensive visual grounding frameworks, driving the development of more reliable and robust multimodal systems for complex real-world scenarios.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 19:43:44 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Jin",
"Guanghao",
""
],
[
"Wu",
"Jingpei",
""
],
[
"Guo",
"Tianpei",
""
],
[
"Niu",
"Yiyi",
""
],
[
"Zhou",
"Weidong",
""
],
[
"Liu",
"Guoyang",
""
]
] |
2508.14081
|
Yoshimasa Kubo
|
Yoshimasa Kubo and Jean Erik Delanois and Maxim Bazhenov
|
Toward Lifelong Learning in Equilibrium Propagation: Sleep-like and Awake Rehearsal for Enhanced Stability
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Recurrent neural networks (RNNs) trained using Equilibrium Propagation (EP), a biologically plausible training algorithm, have demonstrated strong performance in various tasks such as image classification and reinforcement learning. However, these networks face a critical challenge in continuous learning: catastrophic forgetting, where previously acquired knowledge is overwritten when new tasks are learned. This limitation contrasts with the human brain's ability to retain and integrate both old and new knowledge, aided by processes like memory consolidation during sleep through the replay of learned information. To address this challenge in RNNs, here we propose a sleep-like replay consolidation (SRC) algorithm for EP-trained RNNs. We found that SRC significantly improves RNN's resilience to catastrophic forgetting in continuous learning scenarios. In class-incremental learning with SRC implemented after each new task training, the EP-trained multilayer RNN model (MRNN-EP) performed significantly better compared to feedforward networks incorporating several well-established regularization techniques. The MRNN-EP performed on par with MRNN trained using Backpropagation Through Time (BPTT) when both were equipped with SRC on MNIST data and surpassed BPTT-based models on the Fashion MNIST, Kuzushiji-MNIST, CIFAR10, and ImageNet datasets. Combining SRC with rehearsal, also known as "awake replay", further boosted the network's ability to retain long-term knowledge while continuing to learn new tasks. Our study reveals the applicability of sleep-like replay techniques to RNNs and highlights the potential for integrating human-like learning behaviors into artificial neural networks (ANNs).
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 20:48:02 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kubo",
"Yoshimasa",
""
],
[
"Delanois",
"Jean Erik",
""
],
[
"Bazhenov",
"Maxim",
""
]
] |
2508.14082
|
Hezhe Qiao
|
Ye Su, Hezhe Qiao, Wei Huang, Lin Chen
|
Toward Generalist Semi-supervised Regression via Decoupled Representation Distillation
|
12 pages
| null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Semi-supervised regression (SSR), which aims to predict continuous scores of samples while reducing reliance on a large amount of labeled data, has recently received considerable attention across various applications, including computer vision, natural language processing, and audio and medical analysis. Existing semi-supervised methods typically apply consistency regularization on the general regression task by generating pseudo-labels. However, these methods heavily rely on the quality of pseudo-labels, and direct regression fails to learn the label distribution and can easily lead to overfitting. To address these challenges, we introduce an end-to-end Decoupled Representation distillation framework (DRILL) which is specially designed for the semi-supervised regression task where we transform the general regression task into a Discrete Distribution Estimation (DDE) task over multiple buckets to better capture the underlying label distribution and mitigate the risk of overfitting associated with direct regression. Then we employ the Decoupled Distribution Alignment (DDA) to align the target bucket and non-target bucket between teacher and student on the distribution of buckets, encouraging the student to learn more robust and generalized knowledge from the teacher. Extensive experiments conducted on datasets from diverse domains demonstrate that the proposed DRILL has strong generalization and outperforms the competing methods.
|
[
{
"version": "v1",
"created": "Tue, 12 Aug 2025 22:11:01 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Su",
"Ye",
""
],
[
"Qiao",
"Hezhe",
""
],
[
"Huang",
"Wei",
""
],
[
"Chen",
"Lin",
""
]
] |
2508.14083
|
Songyu Ke
|
Songyu Ke and Chenyu Wu and Yuxuan Liang and Xiuwen Yi and Yanping Sun and Junbo Zhang and Yu Zheng
|
GeoMAE: Masking Representation Learning for Spatio-Temporal Graph Forecasting with Missing Values
|
8 pages, incomplete version
| null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Accurate acquisition of crowd flow at Points of Interest (POIs) is pivotal for effective traffic management, public service, and urban planning. Despite this importance, due to the limitations of urban sensing techniques, the data quality from most sources is inadequate for monitoring crowd flow at each POI. This renders the inference of accurate crowd flow from low-quality data a critical and challenging task. The complexity is heightened by three key factors: 1) \emph{The scarcity and rarity of labeled data}, 2) \emph{The intricate spatio-temporal dependencies among POIs}, and 3) \emph{The myriad correlations between precise crowd flow and GPS reports}.
To address these challenges, we recast the crowd flow inference problem as a self-supervised attributed graph representation learning task and introduce a novel \underline{C}ontrastive \underline{S}elf-learning framework for \underline{S}patio-\underline{T}emporal data (\model). Our approach initiates with the construction of a spatial adjacency graph founded on the POIs and their respective distances. We then employ a contrastive learning technique to exploit large volumes of unlabeled spatio-temporal data. We adopt a swapped prediction approach to anticipate the representation of the target subgraph from similar instances. Following the pre-training phase, the model is fine-tuned with accurate crowd flow data. Our experiments, conducted on two real-world datasets, demonstrate that the \model pre-trained on extensive noisy data consistently outperforms models trained from scratch.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 03:30:45 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ke",
"Songyu",
""
],
[
"Wu",
"Chenyu",
""
],
[
"Liang",
"Yuxuan",
""
],
[
"Yi",
"Xiuwen",
""
],
[
"Sun",
"Yanping",
""
],
[
"Zhang",
"Junbo",
""
],
[
"Zheng",
"Yu",
""
]
] |
2508.14084
|
Minh Chung
|
Minh Chung, Yaknan Gambo, Burak Mete, Xiao-Ting Michelle To, Florian Kr\"otz, Korbinian Staudacher, Martin Letras, Xiaolong Deng, Mounika Vavilala, Amir Raoofy, Jorge Echavarria, Luigi Iapichino, Laura Schulz, Josef Weidendorfer, Martin Schulz
|
Q-BEAST: A Practical Course on Experimental Evaluation and Characterization of Quantum Computing Systems
|
This paper is submitted and accepted in the Fourth Annual Quantum Science and Engineering Education Conference (QSEEC25), which is collocated with the IEEE International Conference on Quantum Computing & Engineering (QCE25), part of IEEE Quantum Week 2025
| null | null | null |
physics.ed-ph cs.ET math.QA quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Quantum computing (QC) promises to be a transformative technology with impact on various application domains, such as optimization, cryptography, and material science. However, the technology has a sharp learning curve, and practical evaluation and characterization of quantum systems remains complex and challenging, particularly for students and newcomers from computer science to the field of quantum computing. To address this educational gap, we introduce Q-BEAST, a practical course designed to provide structured training in the experimental analysis of quantum computing systems. Q-BEAST offers a curriculum that combines foundational concepts in quantum computing with practical methodologies and use cases for benchmarking and performance evaluation on actual quantum systems. Through theoretical instruction and hands-on experimentation, students gain experience in assessing the advantages and limitations of real quantum technologies. With that, Q-BEAST supports the education of a future generation of quantum computing users and developers. Furthermore, it also explicitly promotes a deeper integration of High Performance Computing (HPC) and QC in research and education.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 08:02:05 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Chung",
"Minh",
""
],
[
"Gambo",
"Yaknan",
""
],
[
"Mete",
"Burak",
""
],
[
"To",
"Xiao-Ting Michelle",
""
],
[
"Krötz",
"Florian",
""
],
[
"Staudacher",
"Korbinian",
""
],
[
"Letras",
"Martin",
""
],
[
"Deng",
"Xiaolong",
""
],
[
"Vavilala",
"Mounika",
""
],
[
"Raoofy",
"Amir",
""
],
[
"Echavarria",
"Jorge",
""
],
[
"Iapichino",
"Luigi",
""
],
[
"Schulz",
"Laura",
""
],
[
"Weidendorfer",
"Josef",
""
],
[
"Schulz",
"Martin",
""
]
] |
2508.14085
|
Hanseul Kang
|
Hanseul Kang, Shervin Karimkashi, Ville Vuorinen
|
Parameter-Aware Ensemble SINDy for Interpretable Symbolic SGS Closure
|
16 pages, 6 figures
| null | null | null |
cs.LG physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a scalable, parameter-aware sparse regression framework for discovering interpretable partial differential equations and subgrid-scale closures from multi-parameter simulation data. Building on SINDy (Sparse Identification of Nonlinear Dynamics), our approach addresses key limitations through four innovations: symbolic parameterisation enabling physical parameters to vary within unified regression; Dimensional Similarity Filter enforcing unit-consistency whilst reducing candidate libraries; memory-efficient Gram-matrix accumulation enabling batch processing; and ensemble consensus with coefficient stability analysis for robust model identification.
Validation on canonical one-dimensional benchmarks demonstrates reliable recovery of governing equations across parameter ranges. Applied to filtered Burgers datasets, the framework discovers an SGS closure $\tau_{\mathrm{SGS}} = 0.1603\cdot\Delta^2\left(\frac{\partial \bar{u}}{\partial x}\right)^2$, corresponding to a Smagorinsky constant of approximately 0.4004. This represents autonomous discovery of Smagorinsky-type closure structure from data without prior theoretical assumptions.
The discovered model achieves $R^2 = 0.886$ across filter scales and demonstrates improved prediction accuracy compared to classical closures. The framework's ability to identify physically meaningful SGS forms and calibrate coefficients offers a complementary approach to existing turbulence modelling methods, contributing to the growing field of data-driven closure discovery.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 09:37:57 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kang",
"Hanseul",
""
],
[
"Karimkashi",
"Shervin",
""
],
[
"Vuorinen",
"Ville",
""
]
] |
2508.14086
|
Sim Kuan Goh
|
Jia Hong Puah, Sim Kuan Goh, Ziwei Zhang, Zixuan Ye, Chow Khuen Chan, Kheng Seang Lim, Si Lei Fong, Kok Sin Woon
|
EEGDM: EEG Representation Learning via Generative Diffusion Model
|
EEGDM Preprint
| null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
While electroencephalogram (EEG) has been a crucial tool for monitoring the brain and diagnosing neurological disorders (e.g., epilepsy), learning meaningful representations from raw EEG signals remains challenging due to limited annotations and high signal variability. Recently, EEG foundation models (FMs) have shown promising potential by adopting transformer architectures and self-supervised pre-training methods from large language models (e.g., masked prediction) to learn representations from diverse EEG data, followed by fine-tuning on specific EEG tasks. Nonetheless, these large models often incurred high computational costs during both training and inference, with only marginal performance improvements as model size increases. In this work, we proposed EEG representation learning framework building upon Generative Diffusion Model (EEGDM). Specifically, we developed structured state-space model for diffusion pretraining (SSMDP) to better capture the temporal dynamics of EEG signals and trained the architecture using a Denoising Diffusion Probabilistic Model. The resulting latent EEG representations were then used for downstream classification tasks via our proposed latent fusion transformer (LFT). To evaluate our method, we used the multi-event Temple University EEG Event Corpus and compared EEGDM with current state-of-the-art approaches, including EEG FMs. Empirical results showed that our method outperformed existing methods while being approximately 19x more lightweight. These findings suggested that EEGDM offered a promising alternative to current FMs. Our code is available at: https://github.com/jhpuah/EEGDM.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 14:40:52 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Puah",
"Jia Hong",
""
],
[
"Goh",
"Sim Kuan",
""
],
[
"Zhang",
"Ziwei",
""
],
[
"Ye",
"Zixuan",
""
],
[
"Chan",
"Chow Khuen",
""
],
[
"Lim",
"Kheng Seang",
""
],
[
"Fong",
"Si Lei",
""
],
[
"Woon",
"Kok Sin",
""
]
] |
2508.14087
|
Yihui Ren
|
David Park, Shuhang Li, Yi Huang, Xihaier Luo, Haiwang Yu, Yeonju Go, Christopher Pinkenburg, Yuewei Lin, Shinjae Yoo, Joseph Osborn, Jin Huang, Yihui Ren
|
FM4NPP: A Scaling Foundation Model for Nuclear and Particle Physics
| null | null | null | null |
cs.LG cs.AI hep-ex
|
http://creativecommons.org/licenses/by/4.0/
|
Large language models have revolutionized artificial intelligence by enabling large, generalizable models trained through self-supervision. This paradigm has inspired the development of scientific foundation models (FMs). However, applying this capability to experimental particle physics is challenging due to the sparse, spatially distributed nature of detector data, which differs dramatically from natural language. This work addresses if an FM for particle physics can scale and generalize across diverse tasks. We introduce a new dataset with more than 11 million particle collision events and a suite of downstream tasks and labeled data for evaluation. We propose a novel self-supervised training method for detector data and demonstrate its neural scalability with models that feature up to 188 million parameters. With frozen weights and task-specific adapters, this FM consistently outperforms baseline models across all downstream tasks. The performance also exhibits robust data-efficient adaptation. Further analysis reveals that the representations extracted by the FM are task-agnostic but can be specialized via a single linear mapping for different downstream tasks.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 15:05:06 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Park",
"David",
""
],
[
"Li",
"Shuhang",
""
],
[
"Huang",
"Yi",
""
],
[
"Luo",
"Xihaier",
""
],
[
"Yu",
"Haiwang",
""
],
[
"Go",
"Yeonju",
""
],
[
"Pinkenburg",
"Christopher",
""
],
[
"Lin",
"Yuewei",
""
],
[
"Yoo",
"Shinjae",
""
],
[
"Osborn",
"Joseph",
""
],
[
"Huang",
"Jin",
""
],
[
"Ren",
"Yihui",
""
]
] |
2508.14088
|
Haomin Wen
|
Haomin Wen, Shurui Cao, Leman Akoglu
|
CoBAD: Modeling Collective Behaviors for Human Mobility Anomaly Detection
| null | null | null | null |
cs.LG cs.AI cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
Detecting anomalies in human mobility is essential for applications such as public safety and urban planning. While traditional anomaly detection methods primarily focus on individual movement patterns (e.g., a child should stay at home at night), collective anomaly detection aims to identify irregularities in collective mobility behaviors across individuals (e.g., a child is at home alone while the parents are elsewhere) and remains an underexplored challenge. Unlike individual anomalies, collective anomalies require modeling spatiotemporal dependencies between individuals, introducing additional complexity. To address this gap, we propose CoBAD, a novel model designed to capture Collective Behaviors for human mobility Anomaly Detection. We first formulate the problem as unsupervised learning over Collective Event Sequences (CES) with a co-occurrence event graph, where CES represents the event sequences of related individuals. CoBAD then employs a two-stage attention mechanism to model both the individual mobility patterns and the interactions across multiple individuals. Pre-trained on large-scale collective behavior data through masked event and link reconstruction tasks, CoBAD is able to detect two types of collective anomalies: unexpected co-occurrence anomalies and absence anomalies, the latter of which has been largely overlooked in prior work. Extensive experiments on large-scale mobility datasets demonstrate that CoBAD significantly outperforms existing anomaly detection baselines, achieving an improvement of 13%-18% in AUCROC and 19%-70% in AUCPR. All source code is available at https://github.com/wenhaomin/CoBAD.
|
[
{
"version": "v1",
"created": "Wed, 13 Aug 2025 19:33:38 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wen",
"Haomin",
""
],
[
"Cao",
"Shurui",
""
],
[
"Akoglu",
"Leman",
""
]
] |
2508.14089
|
Nihar Mahapatra
|
Ishaan Mahapatra and Nihar R. Mahapatra
|
Systematic FAIRness Assessment of Open Voice Biomarker Datasets for Mental Health and Neurodegenerative Diseases
|
To appear in the Proceedings of the 28th International Conference on Text, Speech and Dialogue (TSD 2025), Erlangen, Germany, August 25-28, 2025
| null | null | null |
cs.SD cs.LG eess.AS
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Voice biomarkers--human-generated acoustic signals such as speech, coughing, and breathing--are promising tools for scalable, non-invasive detection and monitoring of mental health and neurodegenerative diseases. Yet, their clinical adoption remains constrained by inconsistent quality and limited usability of publicly available datasets. To address this gap, we present the first systematic FAIR (Findable, Accessible, Interoperable, Reusable) evaluation of 27 publicly available voice biomarker datasets focused on these disease areas. Using the FAIR Data Maturity Model and a structured, priority-weighted scoring method, we assessed FAIRness at subprinciple, principle, and composite levels. Our analysis revealed consistently high Findability but substantial variability and weaknesses in Accessibility, Interoperability, and Reusability. Mental health datasets exhibited greater variability in FAIR scores, while neurodegenerative datasets were slightly more consistent. Repository choice also significantly influenced FAIRness scores. To enhance dataset quality and clinical utility, we recommend adopting structured, domain-specific metadata standards, prioritizing FAIR-compliant repositories, and routinely applying structured FAIR evaluation frameworks. These findings provide actionable guidance to improve dataset interoperability and reuse, thereby accelerating the clinical translation of voice biomarker technologies.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 06:55:27 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Mahapatra",
"Ishaan",
""
],
[
"Mahapatra",
"Nihar R.",
""
]
] |
2508.14090
|
Chen Xu
|
Chen Xu, Dawei Yang
|
DLLMQuant: Quantizing Diffusion-based Large Language Models
|
12 pages, 6 figures
| null | null | null |
cs.CL cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Diffusion-based large language models (DLLMs) have shown promise for non-autoregressive text generation, but their deployment is constrained by large model sizes and heavy computational costs. Post-training quantization (PTQ), a widely used method for compressing and accelerating Large Language Models (LLMs), suffers from severe accuracy degradation and reduced generalization performance when directly applied to DLLMs (e.g., AWQ suffers a 16% accuracy drop on LLADA under W4A4). This paper explores how DLLMs' key mechanisms - dynamic masking, iterative generation, bidirectional attention - clash with quantization. We identify three core issues: 1) Iterative generation and dynamic masking ratios lead to distinct token distributions across decoding steps, which are not adequately captured by existing PTQ calibration methods; 2) Quantization errors are accumulated and amplified progressively during iteration in DLLMs, causing quantized models to perform worse as decoding steps progress; 3) Unmasked tokens stabilize while masked remain probabilistic, making overall feature distribution incompatible with existing PTQ methods. To address these issues, we propose DLLMQuant, a PTQ framework tailored for DLLMs, which incorporates three novel techniques: 1) Temporal-Mask Adaptive Sampling (TMAS), a calibration method that accounts for both time and mask factors, with the capacity to capture distributions across timesteps. 2) Interaction-Aware Activation Quantization (IA-AQ), which utilizes bidirectional attention's interaction signals to dynamically allocate quantization resources. 3) Certainty-Guided Quantization (CGQ), which integrates mask status and token scores as key weighting criteria into error compensation, making weight quantization more suitable for DLLMs. Experiments show that DLLMQuant achieves significant performance gains while enhancing efficiency.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 09:30:17 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Xu",
"Chen",
""
],
[
"Yang",
"Dawei",
""
]
] |
2508.14091
|
Matthew Morris Mr
|
Matthew Morris, David J. Tena Cucala, Bernardo Cuenca Grau
|
Logical Expressivity and Explanations for Monotonic GNNs with Scoring Functions
|
Full version (with appendices) of paper accepted to KR 2025 (22nd International Conference on Principles of Knowledge Representation and Reasoning)
| null | null | null |
cs.LG cs.AI cs.LO
|
http://creativecommons.org/licenses/by/4.0/
|
Graph neural networks (GNNs) are often used for the task of link prediction: predicting missing binary facts in knowledge graphs (KGs). To address the lack of explainability of GNNs on KGs, recent works extract Datalog rules from GNNs with provable correspondence guarantees. The extracted rules can be used to explain the GNN's predictions; furthermore, they can help characterise the expressive power of various GNN models. However, these works address only a form of link prediction based on a restricted, low-expressivity graph encoding/decoding method. In this paper, we consider a more general and popular approach for link prediction where a scoring function is used to decode the GNN output into fact predictions. We show how GNNs and scoring functions can be adapted to be monotonic, use the monotonicity to extract sound rules for explaining predictions, and leverage existing results about the kind of rules that scoring functions can capture. We also define procedures for obtaining equivalent Datalog programs for certain classes of monotonic GNNs with scoring functions. Our experiments show that, on link prediction benchmarks, monotonic GNNs and scoring functions perform well in practice and yield many sound rules.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 15:56:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Morris",
"Matthew",
""
],
[
"Cucala",
"David J. Tena",
""
],
[
"Grau",
"Bernardo Cuenca",
""
]
] |
2508.14092
|
Mahmoud Alsaleh
|
Rowanda Ahmed, Belaynesh Chekol and Mahmoud Alsaleh
|
HDBMS: A Context-Aware Hybrid Graph Traversal Algorithm for Efficient Information Discovery in Social Networks
| null | null | null | null |
cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
Graph-searching algorithms play a crucial role in various computational domains, enabling efficient exploration and pathfinding in structured data. Traditional approaches, such as Depth-First Search (DFS) and Breadth-First Search (BFS), follow rigid traversal patterns -- DFS explores branches exhaustively, while BFS expands level by level. In this paper, we propose the Hybrid Depth-Breadth Meaningful Search (HDBMS) algorithm, a novel graph traversal method that dynamically adapts its exploration strategy based on probabilistic node transitions. Unlike conventional methods, HDBMS prioritizes traversal paths by estimating the likelihood that a node contains the desired information, ensuring a more contextually relevant search. Through extensive experimentation on diverse directed graphs with varying structural properties, we demonstrate that HDBMS not only maintains competitive computational efficiency but also outperforms traditional algorithms in identifying meaningful paths. By integrating probabilistic decision-making, HDBMS constructs an adaptive and structured traversal order that balances exploration across depth and breadth, making it particularly effective in applications such as information retrieval, social network analysis, and recommendation systems. Our results highlight the robustness of HDBMS in scenarios where the most valuable connections emerge unpredictably, positioning it as a powerful alternative to traditional graph-searching techniques.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 18:12:36 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ahmed",
"Rowanda",
""
],
[
"Chekol",
"Belaynesh",
""
],
[
"Alsaleh",
"Mahmoud",
""
]
] |
2508.14093
|
Daniel Ajeleye
|
Daniel Ajeleye, Ashutosh Trivedi, Majid Zamani
|
Physics-Informed Reward Machines
|
20 pages, currently under review in a conference
| null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Reward machines (RMs) provide a structured way to specify non-Markovian rewards in reinforcement learning (RL), thereby improving both expressiveness and programmability. Viewed more broadly, they separate what is known about the environment, captured by the reward mechanism, from what remains unknown and must be discovered through sampling. This separation supports techniques such as counterfactual experience generation and reward shaping, which reduce sample complexity and speed up learning. We introduce physics-informed reward machines (pRMs), a symbolic machine designed to express complex learning objectives and reward structures for RL agents, thereby enabling more programmable, expressive, and efficient learning. We present RL algorithms capable of exploiting pRMs via counterfactual experiences and reward shaping. Our experimental results show that these techniques accelerate reward acquisition during the training phases of RL. We demonstrate the expressiveness and effectiveness of pRMs through experiments in both finite and continuous physical environments, illustrating that incorporating pRMs significantly improves learning efficiency across several control tasks.
|
[
{
"version": "v1",
"created": "Thu, 14 Aug 2025 18:46:54 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ajeleye",
"Daniel",
""
],
[
"Trivedi",
"Ashutosh",
""
],
[
"Zamani",
"Majid",
""
]
] |
2508.14095
|
Isaac Wolford
|
Isaac Wolford
|
Combinatorial and Gaussian Foundations of Rational Nth Root Approximations: Theorems and Conjectures
|
Pages: 28 Total Words: 5579 Headers: 28 Math Inline: 284 Math Display: 55
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We present an approach (the biroot method) for nth root approximation that yields closed-form rational functions with coefficients derived from binomial structures, Gaussian functions, or qualifying DAG structures. The method emerges from an analysis of Newton's method applied to root extraction, revealing that successive iterations generate coefficients following rows of Pascal's triangle in an alternating numerator-denominator pattern. After further exploration of these patterns, we formulate three main conjectures: (1) the Binomial Biroot Conjecture establishing the fundamental alternating coefficient structure to approximate nth roots (for which we prove the square root case and optimal parameter conditions), (2) the Gaussian Biroot Conjecture, and (3) the General DAG Biroot Conjecture showing a structural invariance to nth root approximation using arbitrary linearly-constructed directed acyclic graphs (DAGs). Computational evidence demonstrates superior convergence properties when compared to Taylor series and Pad\'e approximations, especially considering the more direct and less computationally intensive approach to the biroot function construction. A computational framework has been developed to support systematic exploration of the biroot method's parameter space and to enable extensive numerical and symbolic analysis. The method provides both theoretical insights and computational significance by connecting combinatorial structures to nth root rational approximation theory.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 01:49:52 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wolford",
"Isaac",
""
]
] |
2508.14096
|
Wei Junhao
|
Zhanxi Xie, Baili Lu, Yanzhao Gu, Zikun Li, Junhao Wei and Ngai Cheong
|
Research on UAV Applications in Public Administration: Based on an Improved RRT Algorithm
| null | null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
This study investigates the application of unmanned aerial vehicles (UAVs) in public management, focusing on optimizing path planning to address challenges such as energy consumption, obstacle avoidance, and airspace constraints. As UAVs transition from 'technical tools' to 'governance infrastructure', driven by advancements in low-altitude economy policies and smart city demands, efficient path planning becomes critical. The research proposes an enhanced Rapidly-exploring Random Tree algorithm (dRRT), incorporating four strategies: Target Bias (to accelerate convergence), Dynamic Step Size (to balance exploration and obstacle navigation), Detour Priority (to prioritize horizontal detours over vertical ascents), and B-spline smoothing (to enhance path smoothness). Simulations in a 500 m3 urban environment with randomized buildings demonstrate dRRT's superiority over traditional RRT, A*, and Ant Colony Optimization (ACO). Results show dRRT achieves a 100\% success rate with an average runtime of 0.01468s, shorter path lengths, fewer waypoints, and smoother trajectories (maximum yaw angles <45{\deg}). Despite improvements, limitations include increased computational overhead from added mechanisms and potential local optima due to goal biasing. The study highlights dRRT's potential for efficient UAV deployment in public management scenarios like emergency response and traffic monitoring, while underscoring the need for integration with real-time obstacle avoidance frameworks. This work contributes to interdisciplinary advancements in urban governance, robotics, and computational optimization.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 04:30:57 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Xie",
"Zhanxi",
""
],
[
"Lu",
"Baili",
""
],
[
"Gu",
"Yanzhao",
""
],
[
"Li",
"Zikun",
""
],
[
"Wei",
"Junhao",
""
],
[
"Cheong",
"Ngai",
""
]
] |
2508.14097
|
William Leeney
|
William Leeney, Alessio Gravina and Davide Bacciu
|
Non-Dissipative Graph Propagation for Non-Local Community Detection
|
Accepted at IJCNN 2025
| null | null | null |
cs.SI cs.AI cs.LG
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Community detection in graphs aims to cluster nodes into meaningful groups, a task particularly challenging in heterophilic graphs, where nodes sharing similarities and membership to the same community are typically distantly connected. This is particularly evident when this task is tackled by graph neural networks, since they rely on an inherently local message passing scheme to learn the node representations that serve to cluster nodes into communities. In this work, we argue that the ability to propagate long-range information during message passing is key to effectively perform community detection in heterophilic graphs. To this end, we introduce the Unsupervised Antisymmetric Graph Neural Network (uAGNN), a novel unsupervised community detection approach leveraging non-dissipative dynamical systems to ensure stability and to propagate long-range information effectively. By employing antisymmetric weight matrices, uAGNN captures both local and global graph structures, overcoming the limitations posed by heterophilic scenarios. Extensive experiments across ten datasets demonstrate uAGNN's superior performance in high and medium heterophilic settings, where traditional methods fail to exploit long-range dependencies. These results highlight uAGNN's potential as a powerful tool for unsupervised community detection in diverse graph environments.
|
[
{
"version": "v1",
"created": "Fri, 15 Aug 2025 12:26:48 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Leeney",
"William",
""
],
[
"Gravina",
"Alessio",
""
],
[
"Bacciu",
"Davide",
""
]
] |
2508.14098
|
Pranay Dugar
|
Pranay Dugar, Mohitvishnu S. Gadde, Jonah Siekmann, Yesh Godse, Aayam Shrestha, Alan Fern
|
No More Marching: Learning Humanoid Locomotion for Short-Range SE(2) Targets
| null | null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Humanoids operating in real-world workspaces must frequently execute task-driven, short-range movements to SE(2) target poses. To be practical, these transitions must be fast, robust, and energy efficient. While learning-based locomotion has made significant progress, most existing methods optimize for velocity-tracking rather than direct pose reaching, resulting in inefficient, marching-style behavior when applied to short-range tasks. In this work, we develop a reinforcement learning approach that directly optimizes humanoid locomotion for SE(2) targets. Central to this approach is a new constellation-based reward function that encourages natural and efficient target-oriented movement. To evaluate performance, we introduce a benchmarking framework that measures energy consumption, time-to-target, and footstep count on a distribution of SE(2) goals. Our results show that the proposed approach consistently outperforms standard methods and enables successful transfer from simulation to hardware, highlighting the importance of targeted reward design for practical short-range humanoid locomotion.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 05:48:52 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Dugar",
"Pranay",
""
],
[
"Gadde",
"Mohitvishnu S.",
""
],
[
"Siekmann",
"Jonah",
""
],
[
"Godse",
"Yesh",
""
],
[
"Shrestha",
"Aayam",
""
],
[
"Fern",
"Alan",
""
]
] |
2508.14099
|
Michal Ciebielski
|
Michal Ciebielski, Victor Dh\'edin, Majid Khadiv
|
Task and Motion Planning for Humanoid Loco-manipulation
| null | null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
This work presents an optimization-based task and motion planning (TAMP) framework that unifies planning for locomotion and manipulation through a shared representation of contact modes. We define symbolic actions as contact mode changes, grounding high-level planning in low-level motion. This enables a unified search that spans task, contact, and motion planning while incorporating whole-body dynamics, as well as all constraints between the robot, the manipulated object, and the environment. Results on a humanoid platform show that our method can generate a broad range of physically consistent loco-manipulation behaviors over long action sequences requiring complex reasoning. To the best of our knowledge, this is the first work that enables the resolution of an integrated TAMP formulation with fully acyclic planning and whole body dynamics with actuation constraints for the humanoid loco-manipulation problem.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 06:45:32 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Ciebielski",
"Michal",
""
],
[
"Dhédin",
"Victor",
""
],
[
"Khadiv",
"Majid",
""
]
] |
2508.14100
|
Nilay Kushawaha
|
Nilay Kushawaha, Carlo Alessi, Lorenzo Fruzzetti, Egidio Falotico
|
Domain Translation of a Soft Robotic Arm using Conditional Cycle Generative Adversarial Network
|
Accepted at IEEE International Conference on Robotic Systems and Applications
| null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Deep learning provides a powerful method for modeling the dynamics of soft robots, offering advantages over traditional analytical approaches that require precise knowledge of the robot's structure, material properties, and other physical characteristics. Given the inherent complexity and non-linearity of these systems, extracting such details can be challenging. The mappings learned in one domain cannot be directly transferred to another domain with different physical properties. This challenge is particularly relevant for soft robots, as their materials gradually degrade over time. In this paper, we introduce a domain translation framework based on a conditional cycle generative adversarial network (CCGAN) to enable knowledge transfer from a source domain to a target domain. Specifically, we employ a dynamic learning approach to adapt a pose controller trained in a standard simulation environment to a domain with tenfold increased viscosity. Our model learns from input pressure signals conditioned on corresponding end-effector positions and orientations in both domains. We evaluate our approach through trajectory-tracking experiments across five distinct shapes and further assess its robustness under noise perturbations and periodicity tests. The results demonstrate that CCGAN-GP effectively facilitates cross-domain skill transfer, paving the way for more adaptable and generalizable soft robotic controllers.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 15:47:35 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Kushawaha",
"Nilay",
""
],
[
"Alessi",
"Carlo",
""
],
[
"Fruzzetti",
"Lorenzo",
""
],
[
"Falotico",
"Egidio",
""
]
] |
2508.14101
|
Akash Choudhuri
|
Akash Choudhuri, Yongjian Zhong, Bijaya Adhikari
|
Implicit Hypergraph Neural Network
|
Submitted to ICDM 2025
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Hypergraphs offer a generalized framework for capturing high-order relationships between entities and have been widely applied in various domains, including healthcare, social networks, and bioinformatics. Hypergraph neural networks, which rely on message-passing between nodes over hyperedges to learn latent representations, have emerged as the method of choice for predictive tasks in many of these domains. These approaches typically perform only a small number of message-passing rounds to learn the representations, which they then utilize for predictions. The small number of message-passing rounds comes at a cost, as the representations only capture local information and forego long-range high-order dependencies. However, as we demonstrate, blindly increasing the message-passing rounds to capture long-range dependency also degrades the performance of hyper-graph neural networks.
Recent works have demonstrated that implicit graph neural networks capture long-range dependencies in standard graphs while maintaining performance. Despite their popularity, prior work has not studied long-range dependency issues on hypergraph neural networks. Here, we first demonstrate that existing hypergraph neural networks lose predictive power when aggregating more information to capture long-range dependency. We then propose Implicit Hypergraph Neural Network (IHNN), a novel framework that jointly learns fixed-point representations for both nodes and hyperedges in an end-to-end manner to alleviate this issue. Leveraging implicit differentiation, we introduce a tractable projected gradient descent approach to train the model efficiently. Extensive experiments on real-world hypergraphs for node classification demonstrate that IHNN outperforms the closest prior works in most settings, establishing a new state-of-the-art in hypergraph learning.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 16:58:59 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Choudhuri",
"Akash",
""
],
[
"Zhong",
"Yongjian",
""
],
[
"Adhikari",
"Bijaya",
""
]
] |
2508.14102
|
Thomas Gallien
|
Thomas Gallien
|
Beyond Fixed Morphologies: Learning Graph Policies with Trust Region Compensation in Variable Action Spaces
| null | null | null | null |
cs.LG cs.RO cs.SY eess.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Trust region-based optimization methods have become foundational reinforcement learning algorithms that offer stability and strong empirical performance in continuous control tasks. Growing interest in scalable and reusable control policies translate also in a demand for morphological generalization, the ability of control policies to cope with different kinematic structures. Graph-based policy architectures provide a natural and effective mechanism to encode such structural differences. However, while these architectures accommodate variable morphologies, the behavior of trust region methods under varying action space dimensionality remains poorly understood. To this end, we conduct a theoretical analysis of trust region-based policy optimization methods, focusing on both Trust Region Policy Optimization (TRPO) and its widely used first-order approximation, Proximal Policy Optimization (PPO). The goal is to demonstrate how varying action space dimensionality influence the optimization landscape, particularly under the constraints imposed by KL-divergence or policy clipping penalties. Complementing the theoretical insights, an empirical evaluation under morphological variation is carried out using the Gymnasium Swimmer environment. This benchmark offers a systematically controlled setting for varying the kinematic structure without altering the underlying task, making it particularly well-suited to study morphological generalization.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 17:29:13 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gallien",
"Thomas",
""
]
] |
2508.14103
|
Ben Gould
|
Ben H. Gould
|
A Morse Mayer-Vietoris Sequence for Cosheaf Homology on Simplicial Complexes
|
Essay submitted for the C3.9 Computational Algebraic Topology course (2025), towards a Masters of Mathematics (MMath) at the University of Oxford
| null | null | null |
math.AT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We investigate the homology of cosheaves over finite simplicial complexes. After constructing the Mayer-Vietoris short exact sequence for this homology theory, we apply discrete Morse theory to this setting, defining the associated Morse chain complex. The main result is the construction of a Morse Mayer-Vietoris short exact sequence of Morse chain complexes, for which we establish a quasi-isomorphism to the standard Mayer-Vietoris sequence. Discussions relating poset-based (co)sheaf definitions to topological ones via Kan extensions and exploring a modern categorical perspective on discrete Morse theory are also included.
|
[
{
"version": "v1",
"created": "Sat, 16 Aug 2025 18:26:08 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Gould",
"Ben H.",
""
]
] |
2508.14104
|
Sirui Hong
|
Yutong Bian, Xianhao Lin, Yupeng Xie, Tianyang Liu, Mingchen Zhuge, Siyuan Lu, Haoming Tang, Jinlin Wang, Jiayi Zhang, Jiaqi Chen, Xiangru Tang, Yongxin Ni, Sirui Hong, Chenglin Wu
|
You Don't Know Until You Click:Automated GUI Testing for Production-Ready Software Evaluation
| null | null | null | null |
cs.SE cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Large Language Models (LLMs) and code agents in software development are rapidly evolving from generating isolated code snippets to producing full-fledged software applications with graphical interfaces, interactive logic, and dynamic behaviors. However, current benchmarks fall short in evaluating such production-ready software, as they often rely on static checks or binary pass/fail scripts, failing to capture the interactive behaviors and runtime dynamics that define real-world usability - qualities that only emerge when an application is actively used. This is the blind spot of current evaluation: you don't know if an app works until you click through it, interact with it, and observe how it responds. To bridge this gap, we introduce RealDevWorld, a novel evaluation framework for automated end-to-end assessment of LLMs' ability to generate production-ready repositories from scratch. It features two key components: (1) RealDevBench, a diverse collection of 194 open-ended software engineering tasks across multiple domains, incorporating multimodal elements to reflect real-world complexity; and (2) AppEvalPilot, a new agent-as-a-judge evaluation system that simulates realistic, GUI-based user interactions to automatically and holistically assess software functional correctness, visual fidelity, and runtime behavior. The framework delivers fine-grained, task-specific diagnostic feedback, supporting nuanced evaluation beyond simple success/failure judgments. Empirical results show that RealDevWorld delivers effective, automatic, and human-aligned evaluations, achieving an accuracy of 0.92 and a correlation of 0.85 with expert human assessments, while significantly reducing the reliance on manual review. This enables scalable, human-aligned assessment of production-level software generated by LLMs. Our code is available on GitHub.
|
[
{
"version": "v1",
"created": "Sun, 17 Aug 2025 07:31:11 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bian",
"Yutong",
""
],
[
"Lin",
"Xianhao",
""
],
[
"Xie",
"Yupeng",
""
],
[
"Liu",
"Tianyang",
""
],
[
"Zhuge",
"Mingchen",
""
],
[
"Lu",
"Siyuan",
""
],
[
"Tang",
"Haoming",
""
],
[
"Wang",
"Jinlin",
""
],
[
"Zhang",
"Jiayi",
""
],
[
"Chen",
"Jiaqi",
""
],
[
"Tang",
"Xiangru",
""
],
[
"Ni",
"Yongxin",
""
],
[
"Hong",
"Sirui",
""
],
[
"Wu",
"Chenglin",
""
]
] |
2508.14105
|
Jahid Chowdhury Choton
|
Jahid Chowdhury Choton, John Woods, William Hsu
|
Efficient Environment Design for Multi-Robot Navigation via Continuous Control
|
12 pages, 3 figures, conference
| null | null | null |
cs.RO
|
http://creativecommons.org/licenses/by/4.0/
|
Multi-robot navigation and path planning in continuous state and action spaces with uncertain environments remains an open challenge. Deep Reinforcement Learning (RL) is one of the most popular paradigms for solving this task, but its real-world application has been limited due to sample inefficiency and long training periods. Moreover, the existing works using RL for multi-robot navigation lack formal guarantees while designing the environment. In this paper, we introduce an efficient and highly customizable environment for continuous-control multi-robot navigation, where the robots must visit a set of regions of interest (ROIs) by following the shortest paths. The task is formally modeled as a Markov Decision Process (MDP). We describe the multi-robot navigation task as an optimization problem and relate it to finding an optimal policy for the MDP. We crafted several variations of the environment and measured the performance using both gradient and non-gradient based RL methods: A2C, PPO, TRPO, TQC, CrossQ and ARS. To show real-world applicability, we deployed our environment to a 3-D agricultural field with uncertainties using the CoppeliaSim robot simulator and measured the robustness by running inference on the learned models. We believe our work will guide the researchers on how to develop MDP-based environments that are applicable to real-world systems and solve them using the existing state-of-the-art RL methods with limited resources and within reasonable time periods.
|
[
{
"version": "v1",
"created": "Sun, 17 Aug 2025 16:18:07 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Choton",
"Jahid Chowdhury",
""
],
[
"Woods",
"John",
""
],
[
"Hsu",
"William",
""
]
] |
2508.14106
|
Surajit Das
|
Surajit Das, Gourav Roy, Pavel Zun
|
High-Throughput Low-Cost Segmentation of Brightfield Microscopy Live Cell Images
| null | null | null | null |
q-bio.QM cs.AI cs.CV eess.IV
|
http://creativecommons.org/licenses/by/4.0/
|
Live cell culture is crucial in biomedical studies for analyzing cell properties and dynamics in vitro. This study focuses on segmenting unstained live cells imaged with bright-field microscopy. While many segmentation approaches exist for microscopic images, none consistently address the challenges of bright-field live-cell imaging with high throughput, where temporal phenotype changes, low contrast, noise, and motion-induced blur from cellular movement remain major obstacles. We developed a low-cost CNN-based pipeline incorporating comparative analysis of frozen encoders within a unified U-Net architecture enhanced with attention mechanisms, instance-aware systems, adaptive loss functions, hard instance retraining, dynamic learning rates, progressive mechanisms to mitigate overfitting, and an ensemble technique. The model was validated on a public dataset featuring diverse live cell variants, showing consistent competitiveness with state-of-the-art methods, achieving 93% test accuracy and an average F1-score of 89% (std. 0.07) on low-contrast, noisy, and blurry images. Notably, the model was trained primarily on bright-field images with limited exposure to phase-contrast microscopy (<10%), yet it generalized effectively to the phase-contrast LIVECell dataset, demonstrating modality, robustness and strong performance. This highlights its potential for real-world laboratory deployment across imaging conditions. The model requires minimal compute power and is adaptable using basic deep learning setups such as Google Colab, making it practical for training on other cell variants. Our pipeline outperforms existing methods in robustness and precision for bright-field microscopy segmentation. The code and dataset are available for reproducibility
|
[
{
"version": "v1",
"created": "Sun, 17 Aug 2025 22:05:58 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Das",
"Surajit",
""
],
[
"Roy",
"Gourav",
""
],
[
"Zun",
"Pavel",
""
]
] |
2508.14107
|
Sujit Roy
|
Sujit Roy, Dinesha V. Hegde, Johannes Schmude, Amy Lin, Vishal Gaur, Rohit Lal, Kshitiz Mandal, Talwinder Singh, Andr\'es Mu\~noz-Jaramillo, Kang Yang, Chetraj Pandey, Jinsu Hong, Berkay Aydin, Ryan McGranaghan, Spiridon Kasapis, Vishal Upendran, Shah Bahauddin, Daniel da Silva, Marcus Freitag, Iksha Gurung, Nikolai Pogorelov, Campbell Watson, Manil Maskey, Juan Bernabe-Moreno, Rahul Ramachandran
|
SuryaBench: Benchmark Dataset for Advancing Machine Learning in Heliophysics and Space Weather Prediction
| null | null | null | null |
astro-ph.SR astro-ph.IM cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
This paper introduces a high resolution, machine learning-ready heliophysics dataset derived from NASA's Solar Dynamics Observatory (SDO), specifically designed to advance machine learning (ML) applications in solar physics and space weather forecasting. The dataset includes processed imagery from the Atmospheric Imaging Assembly (AIA) and Helioseismic and Magnetic Imager (HMI), spanning a solar cycle from May 2010 to July 2024. To ensure suitability for ML tasks, the data has been preprocessed, including correction of spacecraft roll angles, orbital adjustments, exposure normalization, and degradation compensation. We also provide auxiliary application benchmark datasets complementing the core SDO dataset. These provide benchmark applications for central heliophysics and space weather tasks such as active region segmentation, active region emergence forecasting, coronal field extrapolation, solar flare prediction, solar EUV spectra prediction, and solar wind speed estimation. By establishing a unified, standardized data collection, this dataset aims to facilitate benchmarking, enhance reproducibility, and accelerate the development of AI-driven models for critical space weather prediction tasks, bridging gaps between solar physics, machine learning, and operational forecasting.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 00:05:01 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Roy",
"Sujit",
""
],
[
"Hegde",
"Dinesha V.",
""
],
[
"Schmude",
"Johannes",
""
],
[
"Lin",
"Amy",
""
],
[
"Gaur",
"Vishal",
""
],
[
"Lal",
"Rohit",
""
],
[
"Mandal",
"Kshitiz",
""
],
[
"Singh",
"Talwinder",
""
],
[
"Muñoz-Jaramillo",
"Andrés",
""
],
[
"Yang",
"Kang",
""
],
[
"Pandey",
"Chetraj",
""
],
[
"Hong",
"Jinsu",
""
],
[
"Aydin",
"Berkay",
""
],
[
"McGranaghan",
"Ryan",
""
],
[
"Kasapis",
"Spiridon",
""
],
[
"Upendran",
"Vishal",
""
],
[
"Bahauddin",
"Shah",
""
],
[
"da Silva",
"Daniel",
""
],
[
"Freitag",
"Marcus",
""
],
[
"Gurung",
"Iksha",
""
],
[
"Pogorelov",
"Nikolai",
""
],
[
"Watson",
"Campbell",
""
],
[
"Maskey",
"Manil",
""
],
[
"Bernabe-Moreno",
"Juan",
""
],
[
"Ramachandran",
"Rahul",
""
]
] |
2508.14108
|
Satori Tsuzuki Ph.D
|
Satori Tsuzuki
|
Band-Limited Equivalence of Convolution Operators and its Application to Filtered Vorticity Dynamics
| null | null | null | null |
math.NA cs.NA math.FA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In this study, we established a general theorem regarding the equivalence of convolution operators restricted to a finite spectral band. We demonstrated that two kernels with identical Fourier transforms over the resolved band act identically on all band-limited functions, even if their kernels differ outside the band. This property is significant in applied mathematics and computational physics, particularly in scenarios where measurements or simulations are spectrally truncated. As an application, we examine the proportionality relation $S(\boldsymbol {r}) \approx \zeta\,\omega(\boldsymbol{r})$ in filtered vorticity dynamics and clarify why real-space diagnostics can underestimate the spectral proportionality due to unobservable degrees of freedom. Our theoretical findings were supported by numerical illustrations using synthetic data.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 00:07:09 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Tsuzuki",
"Satori",
""
]
] |
2508.14109
|
Shayan Bafandkar
|
Shayan Bafandkar, Sungyong Chung, Homa Khosravian, Alireza Talebpour
|
PAPPL: Personalized AI-Powered Progressive Learning Platform
| null | null | null | null |
cs.CY cs.AI cs.ET
|
http://creativecommons.org/licenses/by/4.0/
|
Engineering education has historically been constrained by rigid, standardized frameworks, often neglecting students' diverse learning needs and interests. While significant advancements have been made in online and personalized education within K-12 and foundational sciences, engineering education at both undergraduate and graduate levels continues to lag in adopting similar innovations. Traditional evaluation methods, such as exams and homework assignments, frequently overlook individual student requirements, impeding personalized educational experiences. To address these limitations, this paper introduces the Personalized AI-Powered Progressive Learning (PAPPL) platform, an advanced Intelligent Tutoring System (ITS) designed specifically for engineering education. It highlights the development of a scalable, data-driven tutoring environment leveraging cutting-edge AI technology to enhance personalized learning across diverse academic disciplines, particularly in STEM fields. PAPPL integrates core ITS components including the expert module, student module, tutor module, and user interface, and utilizes GPT-4o, a sophisticated large language model (LLM), to deliver context-sensitive and pedagogically sound hints based on students' interactions. The system uniquely records student attempts, detects recurring misconceptions, and generates progressively targeted feedback, providing personalized assistance that adapts dynamically to each student's learning profile. Additionally, PAPPL offers instructors detailed analytics, empowering evidence-based adjustments to teaching strategies. This study provides a fundamental framework for the progression of Generative ITSs scalable to all education levels, delivering important perspectives on personalized progressive learning and the wider possibilities of Generative AI in the field of education.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 00:35:24 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Bafandkar",
"Shayan",
""
],
[
"Chung",
"Sungyong",
""
],
[
"Khosravian",
"Homa",
""
],
[
"Talebpour",
"Alireza",
""
]
] |
2508.14110
|
Zhengqian Cheng Mr
|
Zhengqian Cheng and Chris A. Marianetti
|
Mott transition from the non-analyticity of the one-body reduced density-matrix functional
| null | null | null | null |
cond-mat.str-el
|
http://creativecommons.org/licenses/by/4.0/
|
One-body reduced density-matrix functional (1RDMF) theory has yielded promising results for small systems such as molecules, but has not addressed quantum phase transitions such as the Mott transition. Here we explicitly execute the constrained search within a variational ansatz to construct a 1RDMF for the multi-orbital Hubbard model with up to seven orbitals in the thermodynamic limit. The variational ansatz is the \mathcal{N}=3 ansatz of the variational discrete action theory (VDAT), which can be exactly evaluated in d=\infty. The resulting 1RDMF exactly encapsulates the \mathcal{N}=3 VDAT results, which accurately captures Mott and Hund physics. We find that non-analytic behavior emerges in our 1RDMF at fixed integer filling, which gives rise to the Mott transition. We explain this behavior by separating the constrained search into multiple stages, illustrating how a nonzero Hund exchange drives the continuous Mott transition to become first-order. Our approach creates a new path forward for constructing an accurate 1RDMF for strongly correlated electron materials.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 02:23:22 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Cheng",
"Zhengqian",
""
],
[
"Marianetti",
"Chris A.",
""
]
] |
2508.14111
|
Jiaqi Wei
|
Jiaqi Wei, Yuejin Yang, Xiang Zhang, Yuhan Chen, Xiang Zhuang, Zhangyang Gao, Dongzhan Zhou, Guangshuai Wang, Zhiqiang Gao, Juntai Cao, Zijie Qiu, Xuming He, Qiang Zhang, Chenyu You, Shuangjia Zheng, Ning Ding, Wanli Ouyang, Nanqing Dong, Yu Cheng, Siqi Sun, Lei Bai, Bowen Zhou
|
From AI for Science to Agentic Science: A Survey on Autonomous Scientific Discovery
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Artificial intelligence (AI) is reshaping scientific discovery, evolving from specialized computational tools into autonomous research partners. We position Agentic Science as a pivotal stage within the broader AI for Science paradigm, where AI systems progress from partial assistance to full scientific agency. Enabled by large language models (LLMs), multimodal systems, and integrated research platforms, agentic AI shows capabilities in hypothesis generation, experimental design, execution, analysis, and iterative refinement -- behaviors once regarded as uniquely human. This survey provides a domain-oriented review of autonomous scientific discovery across life sciences, chemistry, materials science, and physics. We unify three previously fragmented perspectives -- process-oriented, autonomy-oriented, and mechanism-oriented -- through a comprehensive framework that connects foundational capabilities, core processes, and domain-specific realizations. Building on this framework, we (i) trace the evolution of AI for Science, (ii) identify five core capabilities underpinning scientific agency, (iii) model discovery as a dynamic four-stage workflow, (iv) review applications across the above domains, and (v) synthesize key challenges and future opportunities. This work establishes a domain-oriented synthesis of autonomous scientific discovery and positions Agentic Science as a structured paradigm for advancing AI-driven research.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 05:25:54 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Wei",
"Jiaqi",
""
],
[
"Yang",
"Yuejin",
""
],
[
"Zhang",
"Xiang",
""
],
[
"Chen",
"Yuhan",
""
],
[
"Zhuang",
"Xiang",
""
],
[
"Gao",
"Zhangyang",
""
],
[
"Zhou",
"Dongzhan",
""
],
[
"Wang",
"Guangshuai",
""
],
[
"Gao",
"Zhiqiang",
""
],
[
"Cao",
"Juntai",
""
],
[
"Qiu",
"Zijie",
""
],
[
"He",
"Xuming",
""
],
[
"Zhang",
"Qiang",
""
],
[
"You",
"Chenyu",
""
],
[
"Zheng",
"Shuangjia",
""
],
[
"Ding",
"Ning",
""
],
[
"Ouyang",
"Wanli",
""
],
[
"Dong",
"Nanqing",
""
],
[
"Cheng",
"Yu",
""
],
[
"Sun",
"Siqi",
""
],
[
"Bai",
"Lei",
""
],
[
"Zhou",
"Bowen",
""
]
] |
2508.14113
|
Vinit Hegiste
|
Vinit Hegiste and Vidit Goyal and Tatjana Legler and Martin Ruskowski
|
Federated Action Recognition for Smart Worker Assistance Using FastPose
|
8 pages and submitted to FLTA2025 conference
| null | null | null |
cs.CV cs.AI cs.DC cs.HC
|
http://creativecommons.org/licenses/by/4.0/
|
In smart manufacturing environments, accurate and real-time recognition of worker actions is essential for productivity, safety, and human-machine collaboration. While skeleton-based human activity recognition (HAR) offers robustness to lighting, viewpoint, and background variations, most existing approaches rely on centralized datasets, which are impractical in privacy-sensitive industrial scenarios. This paper presents a federated learning (FL) framework for pose-based HAR using a custom skeletal dataset of eight industrially relevant upper-body gestures, captured from five participants and processed using a modified FastPose model. Two temporal backbones, an LSTM and a Transformer encoder, are trained and evaluated under four paradigms: centralized, local (per-client), FL with weighted federated averaging (FedAvg), and federated ensemble learning (FedEnsemble). On the global test set, the FL Transformer improves over centralized training by +12.4 percentage points, with FedEnsemble delivering a +16.3 percentage points gain. On an unseen external client, FL and FedEnsemble exceed centralized accuracy by +52.6 and +58.3 percentage points, respectively. These results demonstrate that FL not only preserves privacy but also substantially enhances cross-user generalization, establishing it as a practical solution for scalable, privacy-aware HAR in heterogeneous industrial settings.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 09:28:15 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Hegiste",
"Vinit",
""
],
[
"Goyal",
"Vidit",
""
],
[
"Legler",
"Tatjana",
""
],
[
"Ruskowski",
"Martin",
""
]
] |
2508.14114
|
Aditey Nandan
|
Aditey Nandan and Viraj Kumar
|
Ambiguity Resolution with Human Feedback for Code Writing Tasks
|
Accepted at the Proceedings of the 33rd International Conference on Computers in Education (ICCE 2025), Asia-Pacific Society for Computers in Education (APSCE)
| null | null | null |
cs.SE cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Specifications for code writing tasks are usually expressed in natural language and may be ambiguous. Programmers must therefore develop the ability to recognize ambiguities in task specifications and resolve them by asking clarifying questions. We present and evaluate a prototype system, based on a novel technique (ARHF: Ambiguity Resolution with Human Feedback), that (1) suggests specific inputs on which a given task specification may be ambiguous, (2) seeks limited human feedback about the code's desired behavior on those inputs, and (3) uses this feedback to generate code that resolves these ambiguities. We evaluate the efficacy of our prototype, and we discuss the implications of such assistive systems on Computer Science education.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 09:46:26 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Nandan",
"Aditey",
""
],
[
"Kumar",
"Viraj",
""
]
] |
2508.14115
|
Taous Iatariene
|
Taous Iatariene, Alexandre Gu\'erin, Romain Serizel (MULTISPEECH)
|
Towards Low-Latency Tracking of Multiple Speakers With Short-Context Speaker Embeddings
| null |
2025 IEEE 27th International Workshop on Multimedia Signal Processing (MMSP), Sep 2025, Beijin, Chine, China
| null | null |
eess.AS cs.AI cs.SD eess.SP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Speaker embeddings are promising identity-related features that can enhance the identity assignment performance of a tracking system by leveraging its spatial predictions, i.e, by performing identity reassignment. Common speaker embedding extractors usually struggle with short temporal contexts and overlapping speech, which imposes long-term identity reassignment to exploit longer temporal contexts. However, this increases the probability of tracking system errors, which in turn impacts negatively on identity reassignment. To address this, we propose a Knowledge Distillation (KD) based training approach for short context speaker embedding extraction from two speaker mixtures. We leverage the spatial information of the speaker of interest using beamforming to reduce overlap. We study the feasibility of performing identity reassignment over blocks of fixed size, i.e., blockwise identity reassignment, to go towards a low-latency speaker embedding based tracking system. Results demonstrate that our distilled models are effective at short-context embedding extraction and more robust to overlap. Although, blockwise reassignment results indicate that further work is needed to handle simultaneous speech more effectively.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 11:32:13 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Iatariene",
"Taous",
"",
"MULTISPEECH"
],
[
"Guérin",
"Alexandre",
"",
"MULTISPEECH"
],
[
"Serizel",
"Romain",
"",
"MULTISPEECH"
]
] |
2508.14116
|
Mbangula Lameck Amugongo
|
Lameck Mbangula Amugongo, Nicola J Bidwell and Joseph Mwatukange
|
Enriching Moral Perspectives on AI: Concepts of Trust amongst Africans
| null | null | null | null |
cs.CY cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
The trustworthiness of AI is considered essential to the adoption and application of AI systems. However, the meaning of trust varies across industry, research and policy spaces. Studies suggest that professionals who develop and use AI regard an AI system as trustworthy based on their personal experiences and social relations at work. Studies about trust in AI and the constructs that aim to operationalise trust in AI (e.g., consistency, reliability, explainability and accountability). However, the majority of existing studies about trust in AI are situated in Western, Educated, Industrialised, Rich and Democratic (WEIRD) societies. The few studies about trust and AI in Africa do not include the views of people who develop, study or use AI in their work. In this study, we surveyed 157 people with professional and/or educational interests in AI from 25 African countries, to explore how they conceptualised trust in AI. Most respondents had links with workshops about trust and AI in Africa in Namibia and Ghana. Respondents' educational background, transnational mobility, and country of origin influenced their concerns about AI systems. These factors also affected their levels of distrust in certain AI applications and their emphasis on specific principles designed to foster trust. Respondents often expressed that their values are guided by the communities in which they grew up and emphasised communal relations over individual freedoms. They described trust in many ways, including applying nuances of Afro-relationalism to constructs in international discourse, such as reliability and reliance. Thus, our exploratory study motivates more empirical research about the ways trust is practically enacted and experienced in African social realities of AI design, use and governance.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 12:04:40 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Amugongo",
"Lameck Mbangula",
""
],
[
"Bidwell",
"Nicola J",
""
],
[
"Mwatukange",
"Joseph",
""
]
] |
2508.14117
|
Salvatore Cielo
|
Salvatore Cielo, Alexander P\"oppl and Ivan Pribec
|
SYCL for Energy-Efficient Numerical Astrophysics: the case of DPEcho
|
11 pages, 6 figures, 2 tables
|
PECS workshop proceedings at EUROPAR 2025
| null | null |
astro-ph.IM cs.PF
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Energy awareness and efficiency policies are gaining more attention, over pure performance (time-to-solution) Key Performance Indicators (KPIs) when comparing the possibilities offered by accelerated systems. But in a field such as numerical astrophysics, which is struggling with code refactorings for GPUs, viable porting paths have to be shown before first. After summarizing the status and recurring problems of astrophysical code accelerations, we highlight how the field would benefit from portable, vendor-agnostic GPU portings. We then employ the DPEcho SYCL benchmark to compare raw performance and energy efficiency for heterogeneous hardware on a realistic application, with the goal of helping computational astrophysicists and HPC providers make informed decisions on the most suitable hardware. Aside from GPUs showing higher efficiency, we argue on the more informative nature of energy-aware KPIs, in that they convey the specific device performance in a data-driven way. We also present a convenient, flexible and cross-platform energy-measuring pipeline. Finally, we contextualize our results through measures with different compilers, presenting device (at the cores) versus node (at the plug) energy and comparing DPEcho with the High- Performance Linpack (HPL) benchmark.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 12:14:52 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Cielo",
"Salvatore",
""
],
[
"Pöppl",
"Alexander",
""
],
[
"Pribec",
"Ivan",
""
]
] |
2508.14118
|
Jason Granstedt
|
Jason Granstedt, Prabhat Kc, Rucha Deshpande, Victor Garcia, Aldo Badano
|
Hallucinations in medical devices
|
19 pages, 2 figures
| null | null | null |
eess.IV cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Computer methods in medical devices are frequently imperfect and are known to produce errors in clinical or diagnostic tasks. However, when deep learning and data-based approaches yield output that exhibit errors, the devices are frequently said to hallucinate. Drawing from theoretical developments and empirical studies in multiple medical device areas, we introduce a practical and universal definition that denotes hallucinations as a type of error that is plausible and can be either impactful or benign to the task at hand. The definition aims at facilitating the evaluation of medical devices that suffer from hallucinations across product areas. Using examples from imaging and non-imaging applications, we explore how the proposed definition relates to evaluation methodologies and discuss existing approaches for minimizing the prevalence of hallucinations.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 12:31:55 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Granstedt",
"Jason",
""
],
[
"Kc",
"Prabhat",
""
],
[
"Deshpande",
"Rucha",
""
],
[
"Garcia",
"Victor",
""
],
[
"Badano",
"Aldo",
""
]
] |
2508.14120
|
Lin Yuhang
|
Yuhang Lin, Yijia Xie, Jiahong Xie, Yuehao Huang, Ruoyu Wang, Jiajun Lv, Yukai Ma, Xingxing Zuo
|
SimGenHOI: Physically Realistic Whole-Body Humanoid-Object Interaction via Generative Modeling and Reinforcement Learning
| null | null | null | null |
cs.RO cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Generating physically realistic humanoid-object interactions (HOI) is a fundamental challenge in robotics. Existing HOI generation approaches, such as diffusion-based models, often suffer from artifacts such as implausible contacts, penetrations, and unrealistic whole-body actions, which hinder successful execution in physical environments. To address these challenges, we introduce SimGenHOI, a unified framework that combines the strengths of generative modeling and reinforcement learning to produce controllable and physically plausible HOI. Our HOI generative model, based on Diffusion Transformers (DiT), predicts a set of key actions conditioned on text prompts, object geometry, sparse object waypoints, and the initial humanoid pose. These key actions capture essential interaction dynamics and are interpolated into smooth motion trajectories, naturally supporting long-horizon generation. To ensure physical realism, we design a contact-aware whole-body control policy trained with reinforcement learning, which tracks the generated motions while correcting artifacts such as penetration and foot sliding. Furthermore, we introduce a mutual fine-tuning strategy, where the generative model and the control policy iteratively refine each other, improving both motion realism and tracking robustness. Extensive experiments demonstrate that SimGenHOI generates realistic, diverse, and physically plausible humanoid-object interactions, achieving significantly higher tracking success rates in simulation and enabling long-horizon manipulation tasks. Code will be released upon acceptance on our project page: https://xingxingzuo.github.io/simgen_hoi.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 15:20:46 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Lin",
"Yuhang",
""
],
[
"Xie",
"Yijia",
""
],
[
"Xie",
"Jiahong",
""
],
[
"Huang",
"Yuehao",
""
],
[
"Wang",
"Ruoyu",
""
],
[
"Lv",
"Jiajun",
""
],
[
"Ma",
"Yukai",
""
],
[
"Zuo",
"Xingxing",
""
]
] |
2508.14121
|
Amadeusz Miszuda
|
A. Miszuda
|
Off-centre convective zones in mass accreting stellar models
|
Accepted for publication in A&AL
| null | null | null |
astro-ph.SR astro-ph.HE
|
http://creativecommons.org/licenses/by/4.0/
|
We report the physical origin of transient off-centre convective zones (oCZs) that arise in mass accreting stellar models. Using detailed MESA simulations of binary evolution, we find that these oCZs are not numerical artefacts but emerge due to a local increase in density near the retreating edge of the convective core. The density enhancement raises the local opacity, which amplifies the radiative temperature gradient $\nabla_{\rm rad}$. If this gradient surpasses the Ledoux threshold $\nabla_{\rm L}$, defined by both thermal and compositional stratification, the region becomes convectively unstable. The resulting oCZs are detached from the convective core and transient: mixing within the oCZ erases the local gradient in mean molecular weight, leaving a sharp $\nabla_{\mu}$ discontinuity at the boundary, stabilizing the adjacent layers. This mechanism naturally explains the presence and evolution of oCZs, as previously reported in massive interacting stars.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 15:41:42 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Miszuda",
"A.",
""
]
] |
2508.14122
|
Alfonso Bueno-Orovio
|
Jolanta Mozyrska, Marcel Beetz, Luke Melas-Kyriazi, Vicente Grau, Abhirup Banerjee, Alfonso Bueno-Orovio
|
3D Cardiac Anatomy Generation Using Mesh Latent Diffusion Models
| null | null | null | null |
eess.IV cs.CV cs.LG q-bio.TO
|
http://creativecommons.org/licenses/by/4.0/
|
Diffusion models have recently gained immense interest for their generative capabilities, specifically the high quality and diversity of the synthesized data. However, examples of their applications in 3D medical imaging are still scarce, especially in cardiology. Generating diverse realistic cardiac anatomies is crucial for applications such as in silico trials, electromechanical computer simulations, or data augmentations for machine learning models. In this work, we investigate the application of Latent Diffusion Models (LDMs) for generating 3D meshes of human cardiac anatomies. To this end, we propose a novel LDM architecture -- MeshLDM. We apply the proposed model on a dataset of 3D meshes of left ventricular cardiac anatomies from patients with acute myocardial infarction and evaluate its performance in terms of both qualitative and quantitative clinical and 3D mesh reconstruction metrics. The proposed MeshLDM successfully captures characteristics of the cardiac shapes at end-diastolic (relaxation) and end-systolic (contraction) cardiac phases, generating meshes with a 2.4% difference in population mean compared to the gold standard.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 16:53:20 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Mozyrska",
"Jolanta",
""
],
[
"Beetz",
"Marcel",
""
],
[
"Melas-Kyriazi",
"Luke",
""
],
[
"Grau",
"Vicente",
""
],
[
"Banerjee",
"Abhirup",
""
],
[
"Bueno-Orovio",
"Alfonso",
""
]
] |
2508.14123
|
Ankita Sharma
|
Ankita Sharma, YuQi Fu, Vahid Ansari, Rishabh Iyer, Fiona Kuang, Kashish Mistry, Raisa Islam Aishy, Sara Ahmad, Joaquin Matres, Dirk R. Englund, and Joyce K.S. Poon
|
AI Agents for Photonic Integrated Circuit Design Automation
| null | null | null | null |
cs.AR cs.AI physics.app-ph physics.optics
|
http://creativecommons.org/licenses/by/4.0/
|
We present Photonics Intelligent Design and Optimization (PhIDO), a multi-agent framework that converts natural-language photonic integrated circuit (PIC) design requests into layout mask files. We compare 7 reasoning large language models for PhIDO using a testbench of 102 design descriptions that ranged from single devices to 112-component PICs. The success rate for single-device designs was up to 91%. For design queries with less than or equal to 15 components, o1, Gemini-2.5-pro, and Claude Opus 4 achieved the highest end-to-end pass@5 success rates of approximately 57%, with Gemini-2.5-pro requiring the fewest output tokens and lowest cost. The next steps toward autonomous PIC development include standardized knowledge representations, expanded datasets, extended verification, and robotic automation.
|
[
{
"version": "v1",
"created": "Mon, 18 Aug 2025 18:20:32 GMT"
}
] | 2025-08-21T00:00:00 |
[
[
"Sharma",
"Ankita",
""
],
[
"Fu",
"YuQi",
""
],
[
"Ansari",
"Vahid",
""
],
[
"Iyer",
"Rishabh",
""
],
[
"Kuang",
"Fiona",
""
],
[
"Mistry",
"Kashish",
""
],
[
"Aishy",
"Raisa Islam",
""
],
[
"Ahmad",
"Sara",
""
],
[
"Matres",
"Joaquin",
""
],
[
"Englund",
"Dirk R.",
""
],
[
"Poon",
"Joyce K. S.",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.