id
string | submitter
string | authors
string | title
string | comments
string | journal-ref
string | doi
string | report-no
string | categories
string | license
string | abstract
string | versions
list | update_date
timestamp[s] | authors_parsed
list |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
2508.16700
|
Divakar Yadav
|
Deepak Kumar, Divakar Yadav, Yash Patel
|
GPT-OSS-20B: A Comprehensive Deployment-Centric Analysis of OpenAI's Open-Weight Mixture of Experts Model
| null | null | null | null |
cs.AR cs.AI cs.DC cs.PF
|
http://creativecommons.org/licenses/by/4.0/
|
We present a single-GPU (H100, bf16) evaluation of GPT-OSS-20B (Mixture-of-Experts; 20.9B total, approx. 3.61B active) against dense baselines Qwen3-32B and Yi-34B across multiple dimensions. We measure true time-to-first-token (TTFT), full-decode throughput (TPOT), end-to-end latency percentiles, peak VRAM with past key values (PKV) held, and energy via a consistent nvidia-smi-based sampler. At a 2048-token context with 64-token decode, GPT-OSS-20B delivers higher decode throughput and tokens per Joule than dense baselines Qwen3-32B and Yi-34B, while substantially reducing peak VRAM and energy per 1000 generated tokens; its TTFT is higher due to MoE routing overhead. With only 17.3% of parameters active (3.61B of 20.9B), GPT-OSS-20B provides about 31.8% higher decode throughput and 25.8% lower energy per 1000 generated tokens than Qwen3-32B at 2048/64, while using 31.7% less peak VRAM. Normalized by active parameters, GPT-OSS-20B shows markedly stronger per-active-parameter efficiency (APE), underscoring MoE's deployment advantages. We do not evaluate accuracy; this is a deployment-focused study. We release code and consolidated results to enable replication and extension.
|
[
{
"version": "v1",
"created": "Fri, 22 Aug 2025 03:37:27 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 03:40:19 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kumar",
"Deepak",
""
],
[
"Yadav",
"Divakar",
""
],
[
"Patel",
"Yash",
""
]
] |
2508.16839
|
Navin Kumar
|
Shayan Vassef, Soorya Ram Shimegekar, Abhay Goyal, Koustuv Saha, Pi Zonooz, and Navin Kumar
|
Route-and-Execute: Auditable Model-Card Matching and Specialty-Level Deployment
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Clinical workflows are fragmented as a patchwork of scripts and task-specific networks that often handle triage, task selection, and model deployment. These pipelines are rarely streamlined for data science pipeline, reducing efficiency and raising operational costs. Workflows also lack data-driven model identification (from imaging/tabular inputs) and standardized delivery of model outputs. In response, we present a practical, healthcare-first framework that uses a single vision-language model (VLM) in two complementary roles. First (Solution 1), the VLM acts as an aware model-card matcher that routes an incoming image to the appropriate specialist model via a three-stage workflow (modality -> primary abnormality -> model-card id). Checks are provided by (i) stagewise prompts that allow early exit via None/Normal/Other and (ii) a stagewise answer selector that arbitrates between the top-2 candidates at each stage, reducing the chance of an incorrect selection and aligning the workflow with clinical risk tolerance. Second (Solution 2), we fine-tune the VLM on specialty-specific datasets ensuring a single model covers multiple downstream tasks within each specialty, maintaining performance while simplifying deployment. Across gastroenterology, hematology, ophthalmology, and pathology, our single-model deployment matches or approaches specialized baselines.
Compared with pipelines composed of many task-specific agents, this approach shows that one VLM can both decide and do. It may reduce effort by data scientists, shorten monitoring, increase the transparency of model selection (with per-stage justifications), and lower integration overhead.
|
[
{
"version": "v1",
"created": "Fri, 22 Aug 2025 23:34:37 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 17:13:21 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 22:39:41 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vassef",
"Shayan",
""
],
[
"Shimegekar",
"Soorya Ram",
""
],
[
"Goyal",
"Abhay",
""
],
[
"Saha",
"Koustuv",
""
],
[
"Zonooz",
"Pi",
""
],
[
"Kumar",
"Navin",
""
]
] |
2508.16879
|
Xi Chen
|
Tianyu Cai, Xi Chen
|
Inverse problem for fractional Schr\"{o}dinger equations with drift on closed Riemannian manifolds
|
There is a gap in the proof
| null | null | null |
math.AP
|
http://creativecommons.org/publicdomain/zero/1.0/
|
This paper is concerned about the inverse coefficient problems of variable-coefficient fractional Schr\"{o}dinger equations with drift on connected closed Riemannian manifolds. We prove that the knowledge of the underlying equation on any non-empty open subset of the underlying manifold determines the Riemannian metric, the drift and the potential, simultaneously and uniquely, up to a gauge transformation. This paper extends the result in \cite{feizmohammadi2021fractionalanisotropiccalderonproblem} for principal terms. Not only can we retrieve lower order terms, but we are also able to achieve the simultaneous inversion of all terms. The key ingredient is a novel Runge approximation of fractional PDEs on Riemannian manifolds.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 02:33:14 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 01:34:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cai",
"Tianyu",
""
],
[
"Chen",
"Xi",
""
]
] |
2508.16889
|
Hyunjun Kim He
|
Hyunjun Kim, Junwoo Ha, Sangyoon Yu, Haon Park
|
ObjexMT: Objective Extraction and Metacognitive Calibration for LLM-as-a-Judge under Multi-Turn Jailbreaks
| null | null | null | null |
cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
LLM-as-a-Judge (LLMaaJ) now underpins scalable evaluation, yet we lack a decisive test of a judge's qualification: can it recover a conversation's latent objective and know when that inference is trustworthy? LLMs degrade under irrelevant or long context; multi-turn jailbreaks further hide goals across turns. We introduce ObjexMT, a benchmark for objective extraction and metacognition. Given a multi-turn transcript, a model must return a one-sentence base objective and a self-reported confidence. Accuracy is computed via LLM-judge semantic similarity to gold objectives, converted to binary correctness by a single human-aligned threshold calibrated once on N = 100 items ($\tau^*=0.61$). Metacognition is evaluated with ECE, Brier, Wrong-at-High-Conf, and risk-coverage. Across gpt-4.1, claude-sonnet-4, and Qwen3-235B-A22B-FP8 on SafeMTData_Attack600, SafeMTData_1K, MHJ, and CoSafe, claude-sonnet-4 attains the best objective-extraction accuracy (0.515) and calibration (ECE 0.296; Brier 0.324); gpt-4.1 and Qwen3-235B-A22B-FP8 tie at 0.441 but are overconfident (mean confidence $\approx$0.88 vs. accuracy $\approx$0.44; Wrong-at-0.90 $\approx$48-52%). Performance varies by dataset ($\approx$0.167-0.865). ObjexMT thus supplies an actionable test for LLM judges: when objectives are not explicit, judges often misinfer them with high confidence. We recommend exposing objectives when feasible and gating decisions by confidence otherwise. Code and data at https://github.com/hyunjun1121/ObjexMT_dataset.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 03:32:04 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 23:34:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kim",
"Hyunjun",
""
],
[
"Ha",
"Junwoo",
""
],
[
"Yu",
"Sangyoon",
""
],
[
"Park",
"Haon",
""
]
] |
2508.16905
|
Mohsen Sheibanian
|
Mohsen Sheibanian, Pouya Shaeri, Alimohammad Beigi, Ryan T. Woo, Aryan Keluskar
|
Tri-Accel: Curvature-Aware Precision-Adaptive and Memory-Elastic Optimization for Efficient GPU Usage
| null | null | null | null |
cs.LG cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Deep neural networks are increasingly bottlenecked by the cost of optimization, both in terms of GPU memory and compute time. Existing acceleration techniques, such as mixed precision, second-order methods, and batch size scaling, are typically used in isolation. We present Tri-Accel, a unified optimization framework that co-adapts three acceleration strategies along with adaptive parameters during training: (1) Precision-Adaptive Updates that dynamically assign mixed-precision levels to layers based on curvature and gradient variance; (2) Sparse Second-Order Signals that exploit Hessian/Fisher sparsity patterns to guide precision and step size decisions; and (3) Memory-Elastic Batch Scaling that adjusts batch size in real time according to VRAM availability. On CIFAR-10 with ResNet-18 and EfficientNet-B0, Tri-Accel achieves up to 9.9% reduction in training time and 13.3% lower memory usage, while improving accuracy by +1.1 percentage points over FP32 baselines. Tested on CIFAR-10/100, our approach demonstrates adaptive learning behavior, with efficiency gradually improving over the course of training as the system learns to allocate resources more effectively. Compared to static mixed-precision training, Tri-Accel maintains 78.1% accuracy while reducing memory footprint from 0.35GB to 0.31GB on standard hardware. The framework is implemented with custom Triton kernels, whose hardware-aware adaptation enables automatic optimization without manual hyperparameter tuning, making it practical for deployment across diverse computational environments. This work demonstrates how algorithmic adaptivity and hardware awareness can be combined to improve scalability in resource-constrained settings, paving the way for more efficient neural network training on edge devices and cost-sensitive cloud deployments.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 05:38:42 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 01:28:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sheibanian",
"Mohsen",
""
],
[
"Shaeri",
"Pouya",
""
],
[
"Beigi",
"Alimohammad",
""
],
[
"Woo",
"Ryan T.",
""
],
[
"Keluskar",
"Aryan",
""
]
] |
2508.16964
|
Maximilian Stahlhofen
|
Maximilian Stahlhofen
|
NNLO electron structure functions (PDFs) from SCET
|
33 pages, 4 figures
| null | null |
FR-PHENO-2025-009
|
hep-ph hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We calculate the electron structure functions, aka parton distribution functions (PDFs), to NNLO in QED. The calculation is based on the definition of the PDFs in terms of operator matrix elements in soft collinear effective theory (SCET) and directly performed in momentum space. The electron PDFs describe the universal effects of collinear initial state radiation (ISR) off the electron (or positron) in DIS or $e^+ e^-$ collision events. The parton collinear to the electron that enters the hard scattering process governed by the energy scale $Q$ can be a photon or a (anti)fermion with mass $m_f \sim m_e \ll Q$. Our SCET momentum-space calculation confirms earlier results obtained from a QED computation in Mellin space and extends them by taking more than one massive fermion flavor into account. The well-known factorization of sufficiently inclusive cross sections into PDFs with massive fermions and hard partonic cross sections with massless fermions at leading order in $m_e/Q$ is discussed from the SCET perspective. Analogies to the nonperturbative PDFs and collinear factorization in QCD are pointed out. Based on this factorization DGLAP-type resummation of large logarithms of the ratio $m_e/Q$ is straightforward.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 09:32:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Stahlhofen",
"Maximilian",
""
]
] |
2508.17009
|
Wangyu Wu
|
Wangyu Wu, Zhenhong Chen, Xiaowen Ma, Wenqiao Zhang, Xianglin Qiu, Siqi Song, Xiaowei Huang, Fei Ma and Jimin Xiao
|
Contrastive Prompt Clustering for Weakly Supervised Semantic Segmentation
| null | null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Weakly Supervised Semantic Segmentation (WSSS) with image-level labels has gained attention for its cost-effectiveness. Most existing methods emphasize inter-class separation, often neglecting the shared semantics among related categories and lacking fine-grained discrimination. To address this, we propose Contrastive Prompt Clustering (CPC), a novel WSSS framework. CPC exploits Large Language Models (LLMs) to derive category clusters that encode intrinsic inter-class relationships, and further introduces a class-aware patch-level contrastive loss to enforce intra-class consistency and inter-class separation. This hierarchical design leverages clusters as coarse-grained semantic priors while preserving fine-grained boundaries, thereby reducing confusion among visually similar categories. Experiments on PASCAL VOC 2012 and MS COCO 2014 demonstrate that CPC surpasses existing state-of-the-art methods in WSSS.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 12:49:08 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 02:43:50 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wu",
"Wangyu",
""
],
[
"Chen",
"Zhenhong",
""
],
[
"Ma",
"Xiaowen",
""
],
[
"Zhang",
"Wenqiao",
""
],
[
"Qiu",
"Xianglin",
""
],
[
"Song",
"Siqi",
""
],
[
"Huang",
"Xiaowei",
""
],
[
"Ma",
"Fei",
""
],
[
"Xiao",
"Jimin",
""
]
] |
2508.17013
|
The-Anh Vu-Le
|
The-Anh Vu-Le, Jo\~ao Alfredo Cardoso Lamy, Tom\'as Alessi, Ian Chen, Minhyuk Park, Elfarouk Harb, George Chacko, and Tandy Warnow
|
Dense Subgraph Clustering and a New Cluster Ensemble Method
| null | null | null | null |
cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
We propose DSC-Flow-Iter, a new community detection algorithm that is based on iterative extraction of dense subgraphs. Although DSC-Flow-Iter leaves many nodes unclustered, it is competitive with leading methods and has high-precision and low-recall, making it complementary to modularity-based methods that typically have high recall but lower precision. Based on this observation, we introduce a novel cluster ensemble technique that combines DSC-Flow-Iter with modularity-based clustering, to provide improved accuracy. We show that our proposed pipeline, which uses this ensemble technique, outperforms its individual components and improves upon the baseline techniques on a large collection of synthetic networks.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 12:53:55 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 23:28:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vu-Le",
"The-Anh",
""
],
[
"Lamy",
"João Alfredo Cardoso",
""
],
[
"Alessi",
"Tomás",
""
],
[
"Chen",
"Ian",
""
],
[
"Park",
"Minhyuk",
""
],
[
"Harb",
"Elfarouk",
""
],
[
"Chacko",
"George",
""
],
[
"Warnow",
"Tandy",
""
]
] |
2508.17121
|
Zhenliang Gan
|
Zhenliang Gan, Xiaoxiao Hu, Sheng Li, Zhenxing Qian, and Xinpeng Zhang
|
SyncGuard: Robust Audio Watermarking Capable of Countering Desynchronization Attacks
|
Accepted at ECAI 2025
| null | null | null |
cs.CR cs.MM cs.SD
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Audio watermarking has been widely applied in copyright protection and source tracing. However, due to the inherent characteristics of audio signals, watermark localization and resistance to desynchronization attacks remain significant challenges. In this paper, we propose a learning-based scheme named SyncGuard to address these challenges. Specifically, we design a frame-wise broadcast embedding strategy to embed the watermark in arbitrary-length audio, enhancing time-independence and eliminating the need for localization during watermark extraction. To further enhance robustness, we introduce a meticulously designed distortion layer. Additionally, we employ dilated residual blocks in conjunction with dilated gated blocks to effectively capture multi-resolution time-frequency features. Extensive experimental results show that SyncGuard efficiently handles variable-length audio segments, outperforms state-of-the-art methods in robustness against various attacks, and delivers superior auditory quality.
|
[
{
"version": "v1",
"created": "Sat, 23 Aug 2025 19:28:04 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 05:12:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Gan",
"Zhenliang",
""
],
[
"Hu",
"Xiaoxiao",
""
],
[
"Li",
"Sheng",
""
],
[
"Qian",
"Zhenxing",
""
],
[
"Zhang",
"Xinpeng",
""
]
] |
2508.17169
|
Yajat Yadav
|
Yajat Yadav, Patrick Mendoza, Jathin Korrapati
|
ONG: Orthogonal Natural Gradient Descent
|
Publicly available code at https://github.com/yajatyadav/orthogonal-natural-gradient
| null | null | null |
cs.LG cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Orthogonal Gradient Descent (OGD) has emerged as a powerful method for continual learning. However, its Euclidean projections do not leverage the underlying information-geometric structure of the problem, which can lead to suboptimal convergence in learning tasks. To address this, we propose incorporating the natural gradient into OGD and present \textbf{ONG (Orthogonal Natural Gradient Descent)}. ONG preconditions each new task-specific gradient with an efficient EKFAC approximation of the inverse Fisher information matrix, yielding updates that follow the steepest descent direction under a Riemannian metric. To preserve performance on previously learned tasks, ONG projects these natural gradients onto the orthogonal complement of prior tasks' gradients. We provide an initial theoretical justification for this procedure, introduce the Orthogonal Natural Gradient Descent (ONG) algorithm, and present preliminary results on the Permuted and Rotated MNIST benchmarks. Our preliminary results, however, indicate that a naive combination of natural gradients and orthogonal projections can have potential issues. This finding motivates continued future work focused on robustly reconciling these geometric perspectives to develop a continual learning method, establishing a more rigorous theoretical foundation with formal convergence guarantees, and extending empirical validation to large-scale continual learning benchmarks. The anonymized version of our code can be found as the zip file here: https://drive.google.com/drive/folders/11PyU6M8pNgOUB5pwdGORtbnMtD8Shiw_?usp=sharing.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 00:27:23 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 04:34:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yadav",
"Yajat",
""
],
[
"Mendoza",
"Patrick",
""
],
[
"Korrapati",
"Jathin",
""
]
] |
2508.17182
|
Hikaru Tsujimura
|
Hikaru Tsujimura, Arush Tagade
|
LLM Assertiveness can be Mechanistically Decomposed into Emotional and Logical Components
|
This preprint is under review
| null | null | null |
cs.LG cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large Language Models (LLMs) often display overconfidence, presenting information with unwarranted certainty in high-stakes contexts. We investigate the internal basis of this behavior via mechanistic interpretability. Using open-sourced Llama 3.2 models fine-tuned on human annotated assertiveness datasets, we extract residual activations across all layers, and compute similarity metrics to localize assertive representations. Our analysis identifies layers most sensitive to assertiveness contrasts and reveals that high-assertive representations decompose into two orthogonal sub-components of emotional and logical clusters-paralleling the dual-route Elaboration Likelihood Model in Psychology. Steering vectors derived from these sub-components show distinct causal effects: emotional vectors broadly influence prediction accuracy, while logical vectors exert more localized effects. These findings provide mechanistic evidence for the multi-component structure of LLM assertiveness and highlight avenues for mitigating overconfident behavior.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 01:43:48 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 21:27:41 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Tsujimura",
"Hikaru",
""
],
[
"Tagade",
"Arush",
""
]
] |
2508.17243
|
Zicong Tang
|
Zicong Tang, Ziyang Ma, Suqing Wang, Zuchao Li, Lefei Zhang, Hai Zhao, Yun Li, Qianren Wang
|
CoViPAL: Layer-wise Contextualized Visual Token Pruning for Large Vision-Language Models
|
Accepted by EMNLP 2025
| null | null | null |
cs.CV cs.AI cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large Vision-Language Models (LVLMs) process multimodal inputs consisting of text tokens and vision tokens extracted from images or videos. Due to the rich visual information, a single image can generate thousands of vision tokens, leading to high computational costs during the prefilling stage and significant memory overhead during decoding. Existing methods attempt to prune redundant vision tokens, revealing substantial redundancy in visual representations. However, these methods often struggle in shallow layers due to the lack of sufficient contextual information. We argue that many visual tokens are inherently redundant even in shallow layers and can be safely and effectively pruned with appropriate contextual signals. In this work, we propose CoViPAL, a layer-wise contextualized visual token pruning method that employs a Plug-and-Play Pruning Module (PPM) to predict and remove redundant vision tokens before they are processed by the LVLM. The PPM is lightweight, model-agnostic, and operates independently of the LVLM architecture, ensuring seamless integration with various models. Extensive experiments on multiple benchmarks demonstrate that CoViPAL outperforms training-free pruning methods under equal token budgets and surpasses training-based methods with comparable supervision. CoViPAL offers a scalable and efficient solution to improve inference efficiency in LVLMs without compromising accuracy.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 07:47:00 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 07:59:19 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Tang",
"Zicong",
""
],
[
"Ma",
"Ziyang",
""
],
[
"Wang",
"Suqing",
""
],
[
"Li",
"Zuchao",
""
],
[
"Zhang",
"Lefei",
""
],
[
"Zhao",
"Hai",
""
],
[
"Li",
"Yun",
""
],
[
"Wang",
"Qianren",
""
]
] |
2508.17343
|
Abhik Roychoudhury
|
Abhik Roychoudhury
|
Agentic AI for Software: thoughts from Software Engineering community
|
4 pages
| null | null | null |
cs.SE cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
AI agents have recently shown significant promise in software engineering. Much public attention has been transfixed on the topic of code generation from Large Language Models (LLMs) via a prompt. However, software engineering is much more than programming, and AI agents go far beyond instructions given by a prompt.
At the code level, common software tasks include code generation, testing, and program repair. Design level software tasks may include architecture exploration, requirements understanding, and requirements enforcement at the code level. Each of these software tasks involves micro-decisions which can be taken autonomously by an AI agent, aided by program analysis tools. This creates the vision of an AI software engineer, where the AI agent can be seen as a member of a development team.
Conceptually, the key to successfully developing trustworthy agentic AI-based software workflows will be to resolve the core difficulty in software engineering - the deciphering and clarification of developer intent. Specification inference, or deciphering the intent, thus lies at the heart of many software tasks, including software maintenance and program repair. A successful deployment of agentic technology into software engineering would involve making conceptual progress in such intent inference via agents.
Trusting the AI agent becomes a key aspect, as software engineering becomes more automated. Higher automation also leads to higher volume of code being automatically generated, and then integrated into code-bases. Thus to deal with this explosion, an emerging direction is AI-based verification and validation (V & V) of AI generated code. We posit that agentic software workflows in future will include such AIbased V&V.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 12:57:21 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 01:35:22 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 06:22:00 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Roychoudhury",
"Abhik",
""
]
] |
2508.17375
|
Junfang Huang
|
Junfang Huang, Yu Yan, Hongzhi Wang, Yingze Li, Jinghan Lin
|
ForeSight: A Predictive-Scheduling Deterministic Database
|
14 pages, 11 figures
| null | null | null |
cs.DB
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Deterministic databases enable scalable replicated systems by executing transactions in a predetermined order. However, existing designs fail to capture transaction dependencies, leading to insufficient scheduling, high abort rates, and poor resource utilization. By addressing these challenges with lightweight conflict prediction and informed scheduling, we present ForeSight, a high-performance deterministic database system. Our system has three core improvements: (1) We design an Association Sum-Product Network to predict potential transaction conflicts, providing the input for dependency analysis without pre-obtained read/write sets. (2) We enhance the storage engine to integrate multi-version-based optimization, improving the execution process and fallback strategy to boost commit rates and concurrency. (3) We propose a matrix two-pass forward scan algorithm that performs dependency analysis to generate conflict-aware schedules, significantly reducing scheduling overhead. Experimental results on multiple benchmarks show that ForeSight achieves up to 2 times higher throughput on skewed workloads and maintains strong performance under contention, demonstrating that predictive scheduling substantially improves deterministic database scalability.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 14:13:36 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 10:33:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Huang",
"Junfang",
""
],
[
"Yan",
"Yu",
""
],
[
"Wang",
"Hongzhi",
""
],
[
"Li",
"Yingze",
""
],
[
"Lin",
"Jinghan",
""
]
] |
2508.17394
|
Nir Mazor
|
Nir Mazor, Tom Hope
|
Lightweight Joint Optimization of General-Purpose Vision-Language Models and Retrievers for Medical Diagnosis
| null | null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by/4.0/
|
Clinical decision-making often involves interpreting images (e.g., radiology) for making diagnoses. Retrieving relevant visual information from medical literature and hospital records could enhance diagnostic accuracy. In this paper, we develop a model in which a multimodal retriever is jointly optimized with an LVLM for medical diagnosis, unlike standard RAG where LVLM error signal is not propagated down to the retriever. We show that using only general-purpose backbones, with only lightweight fine-tuning, our model is able to achieve competitive results with medically-pretrained models across clinical multi-label classification and visual question answering tasks. In a novel analysis, we additionally find that in many cases different top retrieved images each lead to different predictions for a given target, and that these cases are empirically challenging for all models, even for non-retrieval models. Our joint retrieval optimization significantly improves these challenging cases over standard RAG. However, oracle analysis reveals that while the correct diagnosis is frequently achievable using one of the top retrieved images, in practice there is a large performance gap from the oracle, and rerankers using frontier LVLMs do not close this gap -- leaving ample room for improvement by future methods. Code will be made publicly available.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 15:06:20 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 20:43:36 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mazor",
"Nir",
""
],
[
"Hope",
"Tom",
""
]
] |
2508.17403
|
Yinsong Wang
|
Yinsong Wang, Quan Zeng, Xiao Liu, Yu Ding
|
Mutual Information Surprise: Rethinking Unexpectedness in Autonomous Systems
|
Pre-Submission Version
| null | null | null |
cs.LG stat.AP
|
http://creativecommons.org/licenses/by/4.0/
|
Recent breakthroughs in autonomous experimentation have demonstrated remarkable physical capabilities, yet their cognitive control remains limited--often relying on static heuristics or classical optimization. A core limitation is the absence of a principled mechanism to detect and adapt to the unexpectedness. While traditional surprise measures--such as Shannon or Bayesian Surprise--offer momentary detection of deviation, they fail to capture whether a system is truly learning and adapting. In this work, we introduce Mutual Information Surprise (MIS), a new framework that redefines surprise not as anomaly detection, but as a signal of epistemic growth. MIS quantifies the impact of new observations on mutual information, enabling autonomous systems to reflect on their learning progression. We develop a statistical test sequence to detect meaningful shifts in estimated mutual information and propose a mutual information surprise reaction policy (MISRP) that dynamically governs system behavior through sampling adjustment and process forking. Empirical evaluations--on both synthetic domains and a dynamic pollution map estimation task--show that MISRP-governed strategies significantly outperform classical surprise-based approaches in stability, responsiveness, and predictive accuracy. By shifting surprise from reactive to reflective, MIS offers a path toward more self-aware and adaptive autonomous systems.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 15:20:11 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 15:43:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Yinsong",
""
],
[
"Zeng",
"Quan",
""
],
[
"Liu",
"Xiao",
""
],
[
"Ding",
"Yu",
""
]
] |
2508.17439
|
Zijing Zhao
|
Zijing Zhao, Zhu Xu, Qingchao Chen, Yuxin Peng, Yang Liu
|
Investigating Domain Gaps for Indoor 3D Object Detection
|
Accepted to ACM MM 2025
| null |
10.1145/3746027.3758275
| null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
As a fundamental task for indoor scene understanding, 3D object detection has been extensively studied, and the accuracy on indoor point cloud data has been substantially improved. However, existing researches have been conducted on limited datasets, where the training and testing sets share the same distribution. In this paper, we consider the task of adapting indoor 3D object detectors from one dataset to another, presenting a comprehensive benchmark with ScanNet, SUN RGB-D and 3D Front datasets, as well as our newly proposed large-scale datasets ProcTHOR-OD and ProcFront generated by a 3D simulator. Since indoor point cloud datasets are collected and constructed in different ways, the object detectors are likely to overfit to specific factors within each dataset, such as point cloud quality, bounding box layout and instance features. We conduct experiments across datasets on different adaptation scenarios including synthetic-to-real adaptation, point cloud quality adaptation, layout adaptation and instance feature adaptation, analyzing the impact of different domain gaps on 3D object detectors. We also introduce several approaches to improve adaptation performances, providing baselines for domain adaptive indoor 3D object detection, hoping that future works may propose detectors with stronger generalization ability across domains. Our project homepage can be found in https://jeremyzhao1998.github.io/DAVoteNet-release/.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 16:34:19 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 07:43:03 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhao",
"Zijing",
""
],
[
"Xu",
"Zhu",
""
],
[
"Chen",
"Qingchao",
""
],
[
"Peng",
"Yuxin",
""
],
[
"Liu",
"Yang",
""
]
] |
2508.17440
|
Natalia Berloff
|
Nikita Stroev and Natalia G. Berloff
|
Programmable k-local Ising Machines and all-optical Kolmogorov-Arnold Networks on Photonic Platforms
|
16 pages, 6 figures
| null | null | null |
physics.optics cs.ET cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Photonic computing promises energy-efficient acceleration for optimization and learning, yet discrete combinatorial search and continuous function approximation have largely required distinct devices and control stacks. Here we unify k-local Ising optimization and optical Kolmogorov-Arnold network (KAN) learning on a single photonic platform, establishing a critical convergence point in optical computing. We introduce an SLM-centric primitive that realizes, in one stroke, all-optical k-local Ising interactions and fully optical KAN layers. The key idea is to convert the structural nonlinearity of a nominally linear scatterer into a per-window computational resource by adding a single relay pass through the same spatial light modulator: a folded 4f relay re-images the first Fourier plane onto the SLM so that each selected clique or channel occupies a disjoint window with its own second pass phase patch. Propagation remains linear in the optical field, yet the measured intensity in each window becomes a freely programmable polynomial of the clique sum or projection amplitude. This yields native, per clique k-local couplings without nonlinear media and, in parallel, the many independent univariate nonlinearities required by KAN layers, all trainable with in-situ physical gradients using two frames (forward and adjoint). We outline implementations on spatial photonic Ising machines, injection-locked vertical cavity surface emitting laser (VCSEL) arrays, and Microsoft analog optical computers; in all cases the hardware change is one extra lens and a fold (or an on-chip 4f loop), enabling a minimal overhead, massively parallel route to high-order Ising optimization and trainable, all-optical KAN processing on one platform.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 16:39:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 10:53:28 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Stroev",
"Nikita",
""
],
[
"Berloff",
"Natalia G.",
""
]
] |
2508.17481
|
Asaf Shabtai
|
Priyanka Prakash Surve, Asaf Shabtai, Yuval Elovici
|
SoK: Cybersecurity Assessment of Humanoid Ecosystem
| null | null | null | null |
cs.CR cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Humanoids are progressing toward practical deployment across healthcare, industrial, defense, and service sectors. While typically considered cyber-physical systems (CPSs), their dependence on traditional networked software stacks (e.g., Linux operating systems), robot operating system (ROS) middleware, and over-the-air update channels, creates a distinct security profile that exposes them to vulnerabilities conventional CPS models do not fully address. Prior studies have mainly examined specific threats, such as LiDAR spoofing or adversarial machine learning (AML). This narrow focus overlooks how an attack targeting one component can cascade harm throughout the robot's interconnected systems. We address this gap through a systematization of knowledge (SoK) that takes a comprehensive approach, consolidating fragmented research from robotics, CPS, and network security domains. We introduce a seven-layer security model for humanoid robots, organizing 39 known attacks and 35 defenses across the humanoid ecosystem-from hardware to human-robot interaction. Building on this security model, we develop a quantitative 39x35 attack-defense matrix with risk-weighted scoring, validated through Monte Carlo analysis. We demonstrate our method by evaluating three real-world robots: Pepper, G1 EDU, and Digit. The scoring analysis revealed varying security maturity levels, with scores ranging from 39.9% to 79.5% across the platforms. This work introduces a structured, evidence-based assessment method that enables systematic security evaluation, supports cross-platform benchmarking, and guides prioritization of security investments in humanoid robotics.
|
[
{
"version": "v1",
"created": "Sun, 24 Aug 2025 18:13:33 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 16:04:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Surve",
"Priyanka Prakash",
""
],
[
"Shabtai",
"Asaf",
""
],
[
"Elovici",
"Yuval",
""
]
] |
2508.17602
|
Zachary DeStefano
|
Zachary DeStefano and Bufang Liang
|
Push-1 is PSPACE-complete, and the automated verification of motion planning gadgets
|
Added short addendum on concurrent work and small citation fixes
| null | null | null |
cs.CC cs.CG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Push-1 is one of the simplest abstract frameworks for motion planning; however, the complexity of deciding if a Push-1 problem can be solved was a several-decade-old open question. We resolve the complexity of the motion planning problem Push-1 by showing that it is PSPACE-complete, and we formally verify the correctness of our constructions. Our results build upon a recent work which demonstrated that Push-1F (a variant of Push-1 with fixed blocks) and Push-k for $k \geq 2$ (a variant of Push-1 where the agent can push $k$ blocks at once) are PSPACE-complete and more generally on the motion-planning-though-gadgets framework.
In the process of resolving this open problem, we make two general contributions to the motion planning complexity theory. First, our proof technique extends the standard motion planning framework by assigning the agent a state. This state is preserved when traversing between gadgets but can change when taking transitions in gadgets. Second, we designed and implemented a system, GADGETEER, for computationally verifying the behavior of systems of gadgets. This system is agnostic to the underlying motion planning problem, and allows for formally verifying the correspondence between a low-level construction and a high-level system of gadgets as well as automatically synthesizing gadgets from low-level constructions. In the case of Push-1, we use this system to formally prove that our constructions match our high-level specifications of their behavior. This culminates in the construction and verification of a self-closing door, as deciding reachability in a system of self-closing doors is PSPACE-complete.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 02:03:36 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 22:17:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"DeStefano",
"Zachary",
""
],
[
"Liang",
"Bufang",
""
]
] |
2508.17652
|
Xiaobin Sun
|
Mengyu Cheng, Xiaobin Sun and Yingchao Xie
|
Strong averaging principle for nonautonomous multi-scale SPDEs with fully local monotone and almost periodic coefficients
|
29 pages
| null | null | null |
math.PR math.DS
|
http://creativecommons.org/licenses/by/4.0/
|
In this paper, we consider a class of nonautonomous multi-scale stochastic partial differential equations with fully local monotone coefficients. By introducing the evolution system of measures for time-inhomogeneous Markov semigroups, we study the averaging principle for such kind of system. Specifically, we first prove the slow component in the multi-scale stochastic system converges strongly to the solution of an averaged equation, whose coefficients retain the dependence of the scaling parameter. Furthermore, if the coefficients satisfy uniformly almost periodic conditions, we establish that the slow component converges strongly to the solution of another averaged equation, whose coefficients are independent of the scaling parameter. The main contribution of this paper extends the basic nonautonomous framework investigated by Cheng and Liu in [11] to a fully coupled framework, as well as the autonomous framework explored by Liu et al. in [27] to the more general nonautonomous framework. Additionally, we improve the locally monotone coefficients discussed in [11,27] to the fully local monotone coefficients, thus our results can be applied to a wide range of cases in nonlinear nonautonomous stochastic partial differential equations, such as multi-scale stochastic Cahn-Hilliard-heat equation and multi-scale stochastic liquid-crystal-porous-media equation.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 04:26:53 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 13:55:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cheng",
"Mengyu",
""
],
[
"Sun",
"Xiaobin",
""
],
[
"Xie",
"Yingchao",
""
]
] |
2508.17655
|
Hayato Goto
|
Hayato Goto, Ryo Hidaka, Kosuke Tatsumura
|
Harnessing the edge of chaos for combinatorial optimization
|
14 pages, 4 figures, 3 tables
| null | null | null |
quant-ph cs.ET nlin.CD physics.app-ph physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Nonlinear dynamical systems with continuous variables can be used for solving combinatorial optimization problems with discrete variables. In particular, numerical simulations of them can be used as heuristic algorithms with a desirable property, namely, parallelizability, which allows us to execute them in a massively parallel manner using cutting-edge many-core processors, leading to ultrafast performance. However, the dynamical-system approaches with continuous variables are usually less accurate than conventional approaches with discrete variables such as simulated annealing. To improve the solution accuracy of a representative dynamical system-based algorithm called simulated bifurcation (SB), which was found from classical simulation of a quantum nonlinear oscillator network exhibiting quantum bifurcation, here we generalize it by introducing nonlinear control of individual bifurcation parameters and show that the generalized SB (GSB) can achieve almost 100% success probabilities for some large-scale problems. As a result, the time to solution for a 2,000-variable problem is shortened to 10 ms by a GSB-based machine, which is two orders of magnitude shorter than the best known value, 1.3 s, previously obtained by an SB-based machine. To examine the reason for the ultrahigh performance, we investigated chaos in the GSB changing the nonlinear-control strength and found that the dramatic increase of success probabilities happens near the edge of chaos. That is, the GSB can find a solution with high probability by harnessing the edge of chaos. This finding suggests that dynamical-system approaches to combinatorial optimization will be enhanced by harnessing the edge of chaos, opening a broad possibility to tackle intractable combinatorial optimization problems by nature-inspired approaches.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 04:31:57 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 04:05:59 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 08:07:35 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Goto",
"Hayato",
""
],
[
"Hidaka",
"Ryo",
""
],
[
"Tatsumura",
"Kosuke",
""
]
] |
2508.17672
|
Jouko V\"a\"an\"anen
|
Menachem Magidor and Jouko V\"a\"an\"anen
|
New inner models from second order logics
| null | null | null | null |
math.LO
|
http://creativecommons.org/licenses/by/4.0/
|
We define a new inner model C2(omega) based on the fragment of second order logic in which second order variables range over countable subsets of the domain. We compare C2(omega) to the previously studied inner model C(aa). We argue that C2(omega) appears to be a much bigger inner model than C(aa), although this cannot be literally true in ZFC alone. However, we conjecture that it follows from large cardinal assumptions. For example, assuming large cardinals, C2(omega) contains, for every n, an inner model with n Woodin cardinals, while C(aa) contains, under the same assumption, no inner model with a Woodin cardinal. As to large cardinals in C2(omega), we show that, assuming a Woodin limit of Woodin cardinals, the cardinal omega_1 of V is Mahlo in C2(omega). A stronger result is proved for the combination C2(omega, aa) of C(aa) and C2(omega). We also show that the question whether HOD1, a variant of HOD, is the same as HOD cannot be decided on the basis of ZFC even if we add the assumption that there are supercompact cardinals.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 05:10:37 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 19:26:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Magidor",
"Menachem",
""
],
[
"Väänänen",
"Jouko",
""
]
] |
2508.17690
|
Danny Wang
|
Danny Wang, Ruihong Qiu, Guangdong Bai, Zi Huang
|
Text Meets Topology: Rethinking Out-of-distribution Detection in Text-Rich Networks
|
EMNLP2025 Main
| null | null | null |
cs.CL cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Out-of-distribution (OOD) detection remains challenging in text-rich networks, where textual features intertwine with topological structures. Existing methods primarily address label shifts or rudimentary domain-based splits, overlooking the intricate textual-structural diversity. For example, in social networks, where users represent nodes with textual features (name, bio) while edges indicate friendship status, OOD may stem from the distinct language patterns between bot and normal users. To address this gap, we introduce the TextTopoOOD framework for evaluating detection across diverse OOD scenarios: (1) attribute-level shifts via text augmentations and embedding perturbations; (2) structural shifts through edge rewiring and semantic connections; (3) thematically-guided label shifts; and (4) domain-based divisions. Furthermore, we propose TNT-OOD to model the complex interplay between Text aNd Topology using: 1) a novel cross-attention module to fuse local structure into node-level text representations, and 2) a HyperNetwork to generate node-specific transformation parameters. This aligns topological and semantic features of ID nodes, enhancing ID/OOD distinction across structural and textual shifts. Experiments on 11 datasets across four OOD scenarios demonstrate the nuanced challenge of TextTopoOOD for evaluating OOD detection in text-rich networks.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 05:57:29 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 11:53:15 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Danny",
""
],
[
"Qiu",
"Ruihong",
""
],
[
"Bai",
"Guangdong",
""
],
[
"Huang",
"Zi",
""
]
] |
2508.17728
|
Nisreen Albzour
|
Nisreen Albzour and Sarah S. Lam
|
Segmentation and Classification of Pap Smear Images for Cervical Cancer Detection Using Deep Learning
| null | null | null | null |
cs.CV cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Cervical cancer remains a significant global health concern and a leading cause of cancer-related deaths among women. Early detection through Pap smear tests is essential to reduce mortality rates; however, the manual examination is time consuming and prone to human error. This study proposes a deep learning framework that integrates U-Net for segmentation and a classification model to enhance diagnostic performance. The Herlev Pap Smear Dataset, a publicly available cervical cell dataset, was utilized for training and evaluation. The impact of segmentation on classification performance was evaluated by comparing the model trained on segmented images and another trained on non-segmented images. Experimental results showed that the use of segmented images marginally improved the model performance on precision (about 0.41 percent higher) and F1-score (about 1.30 percent higher), which suggests a slightly more balanced classification performance. While segmentation helps in feature extraction, the results showed that its impact on classification performance appears to be limited. The proposed framework offers a supplemental tool for clinical applications, which may aid pathologists in early diagnosis.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 07:11:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Albzour",
"Nisreen",
""
],
[
"Lam",
"Sarah S.",
""
]
] |
2508.17796
|
Yizhou Peng
|
Changsong Liu, Yizhou Peng, and Eng Siong Chng
|
Zero-shot Context Biasing with Trie-based Decoding using Synthetic Multi-Pronunciation
|
Accepted to APSIPA ASC 2025
| null | null | null |
cs.CL eess.AS
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Contextual automatic speech recognition (ASR) systems allow for recognizing out-of-vocabulary (OOV) words, such as named entities or rare words. However, it remains challenging due to limited training data and ambiguous or inconsistent pronunciations. In this paper, we propose a synthesis-driven multi-pronunciation contextual biasing method that performs zero-shot contextual ASR on a pretrained Whisper model. Specifically, we leverage text-to-speech (TTS) systems to synthesize diverse speech samples containing each target rare word, and then use the pretrained Whisper model to extract multiple predicted pronunciation variants. These variant token sequences are compiled into a prefix-trie, which assigns rewards to beam hypotheses in a shallow-fusion manner during beam-search decoding. Subsequently, any recognized variant is mapped back to the original rare word in the final transcription. The evaluation results on the LibriSpeech dataset show that our method reduces biased-word error rate (B-WER) by 43% on test-clean and 44% on test-other while maintaining unbiased-WER (U-WER) essentially unchanged.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 08:41:52 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 02:20:04 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Liu",
"Changsong",
""
],
[
"Peng",
"Yizhou",
""
],
[
"Chng",
"Eng Siong",
""
]
] |
2508.17799
|
Muhammad Afifurrahman
|
Muhammad Afifurrahman, Fawwaz Fakhrurrozi Hadiputra
|
Upper bounds on the odd graceful chromatic number of graphs
|
23 pages, 3 figures
| null | null | null |
math.CO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We obtain several new upper bounds of the odd graceful chromatic number of a graph $G$, which must be bipartite. Some of our bounds depend only on the number of the vertices of $G$ or the chromatic number of some graphs related to the bipartition of $G$.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 08:44:35 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 10:21:03 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Afifurrahman",
"Muhammad",
""
],
[
"Hadiputra",
"Fawwaz Fakhrurrozi",
""
]
] |
2508.17809
|
Abdullah Sahruri
|
Abdullah Sahruri and Martin Margala
|
TLGLock: A New Approach in Logic Locking Using Key-Driven Charge Recycling in Threshold Logic Gates
|
To appear in the 33rd IFIP/IEEE International Conference on Very Large Scale Integration (VLSI-SoC 2025)
| null | null | null |
cs.CR cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Logic locking remains one of the most promising defenses against hardware piracy, yet current approaches often face challenges in scalability and design overhead. In this paper, we present TLGLock, a new design paradigm that leverages the structural expressiveness of Threshold Logic Gates (TLGs) and the energy efficiency of charge recycling to enforce key-dependent functionality at the gate level. By embedding the key into the gate's weighted logic and utilizing dynamic charge sharing, TLGLock provides a stateless and compact alternative to conventional locking techniques. We implement a complete synthesis-to-locking flow and evaluate it using ISCAS, ITC, and MCNC benchmarks. Results show that TLGLock achieves up to 30% area, 50% delay, and 20% power savings compared to latch-based locking schemes. In comparison with XOR and SFLL-HD methods, TLGLock offers up to 3x higher SAT attack resistance with significantly lower overhead. Furthermore, randomized key-weight experiments demonstrate that TLGLock can reach up to 100% output corruption under incorrect keys, enabling tunable security at minimal cost. These results position TLGLock as a scalable and resilient solution for secure hardware design.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 08:57:36 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sahruri",
"Abdullah",
""
],
[
"Margala",
"Martin",
""
]
] |
2508.17865
|
Petr Dunin-Barkowski
|
Alexander Alexandrov, Boris Bychkov, Petr Dunin-Barkowski, Maxim Kazarian, Sergey Shadrin
|
A new spin on polynomial relations among kappa classes
|
19 pages; several corrections and clarifications
| null | null | null |
math.AG hep-th math-ph math.MP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We prove a recent conjecture of the fourth named author with P. Norbury that states a system of universal polynomial relations among the kappa classes on the moduli spaces of algebraic curves. The proof involves localization and materialization analysis of the spin Gromov-Witten theory of the projective line and is dictated by $\mathbb{Z}_2$-equivariant topological recursion.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 10:18:39 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 18:06:33 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Alexandrov",
"Alexander",
""
],
[
"Bychkov",
"Boris",
""
],
[
"Dunin-Barkowski",
"Petr",
""
],
[
"Kazarian",
"Maxim",
""
],
[
"Shadrin",
"Sergey",
""
]
] |
2508.17920
|
Haoshuo Zhang
|
Haoshuo Zhang, Yufei Bo, Hongwei Zhang and Meixia Tao
|
Prompt-based Multimodal Semantic Communication for Multi-spectral Image Segmentation
|
The full-length version, arXiv:2508.20057, has been updated
| null | null | null |
eess.IV cs.MM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Multimodal semantic communication has gained widespread attention due to its ability to enhance downstream task performance. A key challenge in such systems is the effective fusion of features from different modalities, which requires the extraction of rich and diverse semantic representations from each modality. To this end, we propose ProMSC-MIS, a Prompt-based Multimodal Semantic Communication system for Multi-spectral Image Segmentation. Specifically, we propose a pre-training algorithm where features from one modality serve as prompts for another, guiding unimodal semantic encoders to learn diverse and complementary semantic representations. We further introduce a semantic fusion module that combines cross-attention mechanisms and squeeze-and-excitation (SE) networks to effectively fuse cross-modal features. Simulation results show that ProMSC-MIS significantly outperforms benchmark methods across various channel-source compression levels, while maintaining low computational complexity and storage overhead. Our scheme has great potential for applications such as autonomous driving and nighttime surveillance.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 11:38:50 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 11:33:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhang",
"Haoshuo",
""
],
[
"Bo",
"Yufei",
""
],
[
"Zhang",
"Hongwei",
""
],
[
"Tao",
"Meixia",
""
]
] |
2508.17982
|
Prabakaran Rajamanickam
|
Prabakaran Rajamanickam
|
Solute dispersion in axially strained tube flows: Large-time asymptotics and Ornstein-Uhlenbeck Gaussian profiles
|
The asymptotic formula do not satisfy the surface boundary condition
| null | null | null |
physics.flu-dyn
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The dispersion of a passive scalar in an axially strained flow in a slender tube is studied, with particular focus on large-time asymptotics following the approach of~\cite{rajamanickam2020dispersion}. For times exceeding the cross-sectional diffusion timescale, the scalar field forms an axial (Ornstein--Uhlenbeck) Gaussian profile whose variance increases exponentially with the local axial strain, which itself varies with radial location, while radial diffusion only slowly modulates the overall amplitude. In other words, the scalar is dominated by strong axial stretching, completely overwhelming radial diffusion. In striking contrast to classical Taylor dispersion, where radial diffusion rapidly homogenizes the profile and axial convection appears only as a small correction, here axial stretching governs the dominant dynamics, producing a fundamentally different transport mechanism.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 12:48:50 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 09:38:31 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 17:10:17 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Rajamanickam",
"Prabakaran",
""
]
] |
2508.17989
|
Witten Zhao
|
Junwen Zhao, Xue Meng, Wei Zhu, and Congjun Wu
|
Incompressible quantum liquid on the four-dimensional sphere
|
7 pages,3 figures
| null | null | null |
cond-mat.str-el cond-mat.other cond-mat.quant-gas hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The study of quantum Hall effect (QHE) is a foundation of topological physics, inspiring extensive explorations of its high-dimensional generalizations. Notably, the four dimensional (4D) QHE has been experimentally realized in synthetic quantum systems, including cold atoms, photonic lattices, and metamaterials. However, the many-body effect in the 4D QHE system remains poorly understood. In this study, we explore this problem by formulating the microscopic wavefunctions inspired by Laughlin's seminal work. Employing a generalized pseudo-potential framework, we derive an exact microscopic Hamiltonian consisting of two-body projectors that annihilate the microscopic wavefunctions. Diagonalizations on a small size system show that the quasi-hole states remain zero energy while the quasi-particle states exhibit a finite gap, in consistency with an incompressible state. Furthermore, the pairing distribution is calculated to substantiate the liquid-like nature of the wavefunction. Our work provides a preliminary understanding to the fractional topological states in high dimension.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 13:00:20 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 14:51:20 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhao",
"Junwen",
""
],
[
"Meng",
"Xue",
""
],
[
"Zhu",
"Wei",
""
],
[
"Wu",
"Congjun",
""
]
] |
2508.18036
|
Rui Loja Fernandes
|
Rui Loja Fernandes and Ning Jiang
|
Analytic Linearization and Holomorphic Extensions of Proper Groupoids
|
26 pages. Several annoying typos and Example 5.2.4 were corrected
| null | null | null |
math.DG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We establish analytic linearization of s-proper analytic groupoids around invariant submanifolds. We apply this result to show that any such groupoid admits a holomorphic extension.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 13:54:42 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 22:40:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fernandes",
"Rui Loja",
""
],
[
"Jiang",
"Ning",
""
]
] |
2508.18057
|
Wenqiang Sun
|
Wenqiang Sun, Han Yin, Jisheng Bai, Jianfeng Chen
|
Dynamic Fusion Multimodal Network for SpeechWellness Detection
|
6 pages, 5figures
| null | null |
Paper #534
|
cs.SD cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Suicide is one of the leading causes of death among adolescents. Previous suicide risk prediction studies have primarily focused on either textual or acoustic information in isolation, the integration of multimodal signals, such as speech and text, offers a more comprehensive understanding of an individual's mental state. Motivated by this, and in the context of the 1st SpeechWellness detection challenge, we explore a lightweight multi-branch multimodal system based on a dynamic fusion mechanism for speechwellness detection. To address the limitation of prior approaches that rely on time-domain waveforms for acoustic analysis, our system incorporates both time-domain and time-frequency (TF) domain acoustic features, as well as semantic representations. In addition, we introduce a dynamic fusion block to adaptively integrate information from different modalities. Specifically, it applies learnable weights to each modality during the fusion process, enabling the model to adjust the contribution of each modality. To enhance computational efficiency, we design a lightweight structure by simplifying the original baseline model. Experimental results demonstrate that the proposed system exhibits superior performance compared to the challenge baseline, achieving a 78% reduction in model parameters and a 5% improvement in accuracy.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 14:18:12 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 11:20:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Sun",
"Wenqiang",
""
],
[
"Yin",
"Han",
""
],
[
"Bai",
"Jisheng",
""
],
[
"Chen",
"Jianfeng",
""
]
] |
2508.18071
|
Zhenyang Li
|
Zhenyang Li, Xiaoyang Bai, Jinfan Lu, Pengfei Shen, Edmund Y. Lam, Yifan Peng
|
EventTracer: Fast Path Tracing-based Event Stream Rendering
|
15 pages, 7 figures
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Simulating event streams from 3D scenes has become a common practice in event-based vision research, as it meets the demand for large-scale, high temporal frequency data without setting up expensive hardware devices or undertaking extensive data collections. Yet existing methods in this direction typically work with noiseless RGB frames that are costly to render, and therefore they can only achieve a temporal resolution equivalent to 100-300 FPS, far lower than that of real-world event data. In this work, we propose EventTracer, a path tracing-based rendering pipeline that simulates high-fidelity event sequences from complex 3D scenes in an efficient and physics-aware manner. Specifically, we speed up the rendering process via low sample-per-pixel (SPP) path tracing, and train a lightweight event spiking network to denoise the resulting RGB videos into realistic event sequences. To capture the physical properties of event streams, the network is equipped with a bipolar leaky integrate-and-fired (BiLIF) spiking unit and trained with a bidirectional earth mover distance (EMD) loss. Our EventTracer pipeline runs at a speed of about 4 minutes per second of 720p video, and it inherits the merit of accurate spatiotemporal modeling from its path tracing backbone. We show in two downstream tasks that EventTracer captures better scene details and demonstrates a greater similarity to real-world event data than other event simulators, which establishes it as a promising tool for creating large-scale event-RGB datasets at a low cost, narrowing the sim-to-real gap in event-based vision, and boosting various application scenarios such as robotics, autonomous driving, and VRAR.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 14:33:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 16:05:32 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"Zhenyang",
""
],
[
"Bai",
"Xiaoyang",
""
],
[
"Lu",
"Jinfan",
""
],
[
"Shen",
"Pengfei",
""
],
[
"Lam",
"Edmund Y.",
""
],
[
"Peng",
"Yifan",
""
]
] |
2508.18086
|
Achraf Khoudiri
|
Achraf Khoudiri, Khadija El Anouz and Abderrahim El Allati
|
Correlation Enhanced Autonomous Quantum Battery Charging via Structured Reservoirs
|
This work has been submitted as a preprint on arXiv and is intended for future publication in a peer-reviewed journal. The authors retain all rights for formal publication
| null | null | null |
quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We study the charging dynamics of a quantum battery coupled to a structured reservoir composed of two qubits, each in thermal equilibrium with its own bosonic bath. The structured reservoir interacts with a charger battery model through three distinct configurations: (I) direct coupling between the reservoir qubits and the battery, (II) common coupling between the reservoir qubits, the charger, and the battery, and (III) common coupling between the reservoir qubits and the charger, together with a local charger battery interaction. For both incoherent and coherent initial states, we analyze the stored energy, ergotropy, and charging power of the battery, and establish upper and lower bounds on the maximum extractable work of the battery based on the free energy of coherence, and exchanged correlations between the subsystems. Our results show that the nature of the interaction scenario and the presence of coherence crucially determine the efficiency of charging and the maximum ergotropy. We demonstrate that correlation exchange between subsystems, quantified by the difference between global and local coherence, acts as a resource for work extraction. These findings provide a deeper understanding of coherence and population and structured environments as resources for autonomous quantum batteries, and offer experimentally accessible predictions within the regime of superconducting qubits.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 14:46:25 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 23:23:16 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Khoudiri",
"Achraf",
""
],
[
"Anouz",
"Khadija El",
""
],
[
"Allati",
"Abderrahim El",
""
]
] |
2508.18190
|
Zirui Tang
|
Zirui Tang, Boyu Niu, Xuanhe Zhou, Boxiu Li, Wei Zhou, Jiannan Wang, Guoliang Li, Xinyi Zhang, Fan Wu
|
ST-Raptor: LLM-Powered Semi-Structured Table Question Answering
|
Extension of our SIGMOD 2026 paper. Please refer to source code available at: https://github.com/weAIDB/ST-Raptor
| null | null | null |
cs.AI cs.DB cs.IR
|
http://creativecommons.org/licenses/by/4.0/
|
Semi-structured tables, widely used in real-world applications (e.g., financial reports, medical records, transactional orders), often involve flexible and complex layouts (e.g., hierarchical headers and merged cells). These tables generally rely on human analysts to interpret table layouts and answer relevant natural language questions, which is costly and inefficient. To automate the procedure, existing methods face significant challenges. First, methods like NL2SQL require converting semi-structured tables into structured ones, which often causes substantial information loss. Second, methods like NL2Code and multi-modal LLM QA struggle to understand the complex layouts of semi-structured tables and cannot accurately answer corresponding questions. To this end, we propose ST-Raptor, a tree-based framework for semi-structured table question answering using large language models. First, we introduce the Hierarchical Orthogonal Tree (HO-Tree), a structural model that captures complex semi-structured table layouts, along with an effective algorithm for constructing the tree. Second, we define a set of basic tree operations to guide LLMs in executing common QA tasks. Given a user question, ST-Raptor decomposes it into simpler sub-questions, generates corresponding tree operation pipelines, and conducts operation-table alignment for accurate pipeline execution. Third, we incorporate a two-stage verification mechanism: forward validation checks the correctness of execution steps, while backward validation evaluates answer reliability by reconstructing queries from predicted answers. To benchmark the performance, we present SSTQA, a dataset of 764 questions over 102 real-world semi-structured tables. Experiments show that ST-Raptor outperforms nine baselines by up to 20% in answer accuracy. The code is available at https://github.com/weAIDB/ST-Raptor.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 16:48:51 GMT"
},
{
"version": "v2",
"created": "Tue, 26 Aug 2025 08:10:55 GMT"
},
{
"version": "v3",
"created": "Tue, 02 Sep 2025 02:30:46 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Tang",
"Zirui",
""
],
[
"Niu",
"Boyu",
""
],
[
"Zhou",
"Xuanhe",
""
],
[
"Li",
"Boxiu",
""
],
[
"Zhou",
"Wei",
""
],
[
"Wang",
"Jiannan",
""
],
[
"Li",
"Guoliang",
""
],
[
"Zhang",
"Xinyi",
""
],
[
"Wu",
"Fan",
""
]
] |
2508.18204
|
Michael Jasper Martins
|
Jasper Martins, Maria Alessandra Papa, Benjamin Steltner, Reinhard Prix, P. B. Vidal
|
A Bayesian Framework to Follow-up Continuous Gravitational Wave Candidates from Deep Surveys
|
- Updated references - Corrected typos
| null | null | null |
gr-qc astro-ph.IM
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Broad all-sky searches for continuous gravitational waves have high computational costs and require hierarchical pipelines. The sensitivity of these approaches is set by the initial search and by the number of candidates from that stage that can be followed-up. The current follow-up schemes for the deepest surveys require careful tuning and set-up, have a significant human-labor cost and this impacts the number of follow-ups that can be afforded. Here we present and demonstrate a new follow-up framework based on Bayesian parameter estimation for the rapid, highly automated follow-up of candidates produced by the early stages of deep, wide-parameter space searches for continuous waves.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 17:02:56 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 07:51:28 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Martins",
"Jasper",
""
],
[
"Papa",
"Maria Alessandra",
""
],
[
"Steltner",
"Benjamin",
""
],
[
"Prix",
"Reinhard",
""
],
[
"Vidal",
"P. B.",
""
]
] |
2508.18228
|
Donald Stull
|
Marianna Csornyei, D. M. Stull
|
Improved bounds for radial projections in the plane
| null | null | null | null |
math.CA
|
http://creativecommons.org/licenses/by/4.0/
|
We improve the best known lower bound for the dimension of radial projections of sets in the plane. We show that if $X,Y$ are Borel sets in $\R^2$, $X$ is not contained in any line and $\dim_H(X)>0$, then
$$\sup\limits_{x\in X} \dim_H(\pi_x Y) \geq \min\left\{(\dim_H(Y) + \dim_H(X))/2, \dim_H(Y), 1\right\},$$ where $\pi_x Y$ is the radial projection of the set $Y$ from the point $x$.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 17:24:00 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 21:20:45 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Csornyei",
"Marianna",
""
],
[
"Stull",
"D. M.",
""
]
] |
2508.18255
|
Jeffrey Quesnelle
|
Ryan Teknium, Roger Jin, Jai Suphavadeeprasit, Dakota Mahan, Jeffrey Quesnelle, Joe Li, Chen Guang, Shannon Sands, Karan Malhotra
|
Hermes 4 Technical Report
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
We present Hermes 4, a family of hybrid reasoning models that combine structured, multi-turn reasoning with broad instruction-following ability. We describe the challenges encountered during data curation, synthesis, training, and evaluation, and outline the solutions employed to address these challenges at scale. We comprehensively evaluate across mathematical reasoning, coding, knowledge, comprehension, and alignment benchmarks, and we report both quantitative performance and qualitative behavioral analysis. To support open research, all model weights are published publicly at https://huggingface.co/collections/NousResearch/hermes-4-collection-68a731bfd452e20816725728
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 17:45:06 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 17:12:12 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Teknium",
"Ryan",
""
],
[
"Jin",
"Roger",
""
],
[
"Suphavadeeprasit",
"Jai",
""
],
[
"Mahan",
"Dakota",
""
],
[
"Quesnelle",
"Jeffrey",
""
],
[
"Li",
"Joe",
""
],
[
"Guang",
"Chen",
""
],
[
"Sands",
"Shannon",
""
],
[
"Malhotra",
"Karan",
""
]
] |
2508.18334
|
Nelson Colon Vargas
|
Nelson A. Colon Vargas
|
Closed Formulas for $\eta$-Corrections in the Once Punctured Torus
| null | null | null | null |
math.GT math.QA
|
http://creativecommons.org/licenses/by/4.0/
|
We study $\eta$-correction terms in the Kauffman bracket skein algebra of the once-punctured torus $K_t(\Sigma_{1,1})$. While the Frohman-Gelca product-to-sum rule is explicit on the closed torus, punctures introduce correction terms in the ideal $(\eta)$ whose structure has resisted systematic description beyond low-determinant cases. We give a closed form for the family $P_n=(1,0)\cdot T_n((1,2))$: the correction $\epsilon_n$ has an explicit Chebyshev expansion with coefficients that factor as geometric sums in $t^{\pm4}$ governed by parity. Tracking creation and cancellation of $\eta$ through the Chebyshev recurrence, and using diffeomorphism invariance, the formula transports to all products $C_1\cdot T_k(C_2)$ with $|\det(C_1,C_2)|=2$.
We further treat the \emph{primitive maximal-thread} regime, where one Frohman-Gelca summand is fully threaded and the other is simple or doubly covered. In this case we obtain a closed form for the discrepancy: an $\eta$-linear cascade with Chebyshev $S$-coefficients that lowers the thread degree by two at each step. Equivalently, in this maximal-thread regime we solve the specialized Wang-Wong recursion in closed form; the coefficients match their fast algorithm term-by-term (up to the $T_0$ normalization) and subsume Cho's $|\det|=2$ case. The resulting rules give compact, scalable formulas for symbolic multiplication in $K_t(\Sigma_{1,1})$.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 11:57:49 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 12:55:38 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Vargas",
"Nelson A. Colon",
""
]
] |
2508.18338
|
Jarek Duda Dr
|
Jarek Duda, Jagoda Bracha, Adrian Przybysz
|
Linear cost mutual information estimation and independence test of similar performance as HSIC
|
8 pages, 6 figures
| null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Evaluation of statistical dependencies between two data samples is a basic problem of data science/machine learning, and HSIC (Hilbert-Schmidt Information Criterion)~\cite{HSIC} is considered the state-of-art method. However, for size $n$ data sample it requires multiplication of $n\times n$ matrices, what currently needs $\sim O(n^{2.37})$ computational complexity~\cite{mult}, making it impractical for large data samples. We discuss HCR (Hierarchical Correlation Reconstruction) as its linear cost practical alternative, in tests of even higher sensitivity to dependencies, and additionally providing actual joint distribution model for chosen significance level, by description of dependencies through features being mixed moments, starting with correlation and homoscedasticity. Also allowing to approximate mutual information as just sum of squares of such nontrivial mixed moments between two data samples. Such single dependence describing feature is calculated in $O(n)$ linear time. Their number to test varies with dimension $d$ - requiring $O(d^2)$ for pairwise dependencies, $O(d^3)$ if wanting to also consider more subtle triplewise, and so on.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 13:45:25 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 11:45:20 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Duda",
"Jarek",
""
],
[
"Bracha",
"Jagoda",
""
],
[
"Przybysz",
"Adrian",
""
]
] |
2508.18367
|
Ivan Minchev
|
I. Minchev, K. Attard, B. Ratcliffe, M. Martig, J. Walcher, S. Khoperskov, J.P. Bernaldez, L. Marques, K. Sysoliatina, C. Chiappini, M. Steinmetz, R. de Jong
|
The Impact of Radial Migration on Disk Galaxy SFHs: I. Biases in Spatially Resolved Estimates
|
8 pages, 5 figures. Submitted to A&A. Comments welcome!
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Knowledge of the spatially resolved star formation history (SFH) of disk galaxies provides crucial insight into disk assembly, quenching, and chemical evolution. However, most reconstructions, both for the Milky Way and for external galaxies, implicitly assume that stars formed at their present-day radii. Using a range of zoom-in cosmological simulations, we show that stellar radial migration introduces strong and systematic biases in such SFH estimates. In the inner disk (R < h_d), early star formation is typically underestimated by 25-50% and late star formation overestimated, giving the misleading impression of prolonged, moderate activity. An exception occurs in the very central bin considered (~ 0.4h_d), which is consistently overestimated due to a net inflow of inward migrators. At intermediate radii and in the outer disk, migration drives the opposite trend: intermediate-age populations are overestimated by 100-200% as stars born in the inner disk migrate outward, whereas genuinely in-situ populations are underestimated by ~ 50% as they themselves continue to migrate. The net effect is that SFH peaks are suppressed and broadened, and the true rate of inside-out disk growth is systematically underestimated. These distortions affect all galaxies in our sample and have direct implications for interpreting spatially resolved SFHs from IFU surveys such as CALIFA and MaNGA, where present-day radii are often used as proxies for stellar birth sites. Correcting these biases will require accounting for disk mass, bar presence, disk kinematics and morphology, while recent birth-radius estimation techniques for Milky Way stars offer a promising path forward.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 18:00:11 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 14:36:48 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Minchev",
"I.",
""
],
[
"Attard",
"K.",
""
],
[
"Ratcliffe",
"B.",
""
],
[
"Martig",
"M.",
""
],
[
"Walcher",
"J.",
""
],
[
"Khoperskov",
"S.",
""
],
[
"Bernaldez",
"J. P.",
""
],
[
"Marques",
"L.",
""
],
[
"Sysoliatina",
"K.",
""
],
[
"Chiappini",
"C.",
""
],
[
"Steinmetz",
"M.",
""
],
[
"de Jong",
"R.",
""
]
] |
2508.18512
|
Antony C. Chan
|
Antony C Chan
|
Designing across domains with declarative thinking: Insights from the 96-Eyes ptychographic imager project
|
Minor changes: resolve HTML rendering issues of sideways tables; Code listing in dark mode. Cite three more journal articles
| null | null | null |
physics.optics cs.CL
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This article presents a practitioner's reflection on applying declarative, 5th generation, problem formulation language (5GL) to de novo imaging system design, informed by experiences across the interdisciplinary research in academia and cross-functional product development within the private sector. Using the 96-Eyes project: 96-camera parallel multi-modal imager for high-throughput drug discovery as a representative case, I illustrate how project requirements, ranging from hardware constraints to life sciences needs, can be formalized into machine-readable problem statements to preserve mission-critical input from diverse domain stakeholders. This declarative approach enhances transparency, ensures design traceability, and minimizes costly misalignment across optical, algorithmic, hardware-accelerated compute, and life sciences teams.
Alongside the technical discussion of 5GL with real-world code examples, I reflect on the practical barriers to adopting 5GL in environments where imperative, 3rd-generation languages (3GL) remain the default medium for inter-team collaboration. Rather than offering an one-size-fits-all solution, these learned lessons highlight how programming paradigms implicitly shapes research workflows through existing domain hierarchies. The discussion aims to invite further explorations into how declarative problem formulations can facilitate innovation in settings where concurrent R\&{}D workflows are gaining traction, as opposed to environments where sequential, phase-driven workflows remain the norm.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 21:32:15 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 17:31:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chan",
"Antony C",
""
]
] |
2508.18541
|
Jaspreet Ranjit
|
Jaspreet Ranjit, Hyundong J. Cho, Claire J. Smerdon, Yoonsoo Nam, Myles Phung, Jonathan May, John R. Blosnich, Swabha Swayamdipta
|
Uncovering Intervention Opportunities for Suicide Prevention with Language Model Assistants
|
Project Website: https://dill-lab.github.io/interventions_lm_assistants/
| null | null | null |
cs.CY
|
http://creativecommons.org/licenses/by/4.0/
|
Warning: This paper discusses topics of suicide and suicidal ideation, which may be distressing to some readers.
The National Violent Death Reporting System (NVDRS) documents information about suicides in the United States, including free text narratives (e.g., circumstances surrounding a suicide). In a demanding public health data pipeline, annotators manually extract structured information from death investigation records following extensive guidelines developed painstakingly by experts. In this work, we facilitate data-driven insights from the NVDRS data to support the development of novel suicide interventions by investigating the value of language models (LMs) as efficient assistants to these (a) data annotators and (b) experts. We find that LM predictions match existing data annotations about 85% of the time across 50 NVDRS variables. In the cases where the LM disagrees with existing annotations, expert review reveals that LM assistants can surface annotation discrepancies 38% of the time. Finally, we introduce a human-in-the-loop algorithm to assist experts in efficiently building and refining guidelines for annotating new variables by allowing them to focus only on providing feedback for incorrect LM predictions. We apply our algorithm to a real-world case study for a new variable that characterizes victim interactions with lawyers and demonstrate that it achieves comparable annotation quality with a laborious manual approach. Our findings provide evidence that LMs can serve as effective assistants to public health researchers who handle sensitive data in high-stakes scenarios.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 22:30:10 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 21:59:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ranjit",
"Jaspreet",
""
],
[
"Cho",
"Hyundong J.",
""
],
[
"Smerdon",
"Claire J.",
""
],
[
"Nam",
"Yoonsoo",
""
],
[
"Phung",
"Myles",
""
],
[
"May",
"Jonathan",
""
],
[
"Blosnich",
"John R.",
""
],
[
"Swayamdipta",
"Swabha",
""
]
] |
2508.18565
|
Hao Zhou
|
Hao Zhou, Sibo Cheng
|
Improving Long-term Autoregressive Spatiotemporal Predictions: A Proof of Concept with Fluid Dynamics
| null | null |
10.1016/j.cma.2025.118332
| null |
cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Data-driven methods are emerging as efficient alternatives to traditional numerical forecasting, offering fast inference and lower computational cost. Yet, for complex systems, long-term accuracy often deteriorates due to error accumulation, and autoregressive training (though effective) demands large GPU memory and may sacrifice short-term performance. We propose the Stochastic PushForward (SPF) framework, which retains one-step-ahead training while enabling multi-step learning. SPF builds a supplementary dataset from model predictions and combines it with ground truth via a stochastic acquisition strategy, balancing short- and long-term performance while reducing overfitting. Multi-step predictions are precomputed between epochs, keeping memory usage stable without storing full unrolled sequences. Experiments on the Burgers' equation and the Shallow Water benchmark show that SPF achieves higher long-term accuracy than autoregressive methods while lowering memory requirements, making it promising for resource-limited and complex simulations.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 23:51:18 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhou",
"Hao",
""
],
[
"Cheng",
"Sibo",
""
]
] |
2508.18579
|
Ali Motahharynia
|
Mohammadreza Ghaffarzadeh-Esfahani, Ali Motahharynia, Nahid Yousefian, Navid Mazrouei, Jafar Ghaisari, Yousof Gheisari
|
DrugReasoner: Interpretable Drug Approval Prediction with a Reasoning-augmented Language Model
|
13 pages, 2 figures. Corresponding author: alimotahharynia@gmail.com Kaggle notebook: https://www.kaggle.com/code/mohammadgh009/drugreasoner
| null | null | null |
cs.LG cs.AI q-bio.QM
|
http://creativecommons.org/licenses/by-nc-sa/4.0/
|
Drug discovery is a complex and resource-intensive process, making early prediction of approval outcomes critical for optimizing research investments. While classical machine learning and deep learning methods have shown promise in drug approval prediction, their limited interpretability constraints their impact. Here, we present DrugReasoner, a reasoning-based large language model (LLM) built on the LLaMA architecture and fine-tuned with group relative policy optimization (GRPO) to predict the likelihood of small-molecule approval. DrugReasoner integrates molecular descriptors with comparative reasoning against structurally similar approved and unapproved compounds, generating predictions alongside step-by-step rationales and confidence scores. DrugReasoner achieved robust performance with an AUC of 0.732 and an F1 score of 0.729 on the validation set and 0.725 and 0.718 on the test set, respectively. These results outperformed conventional baselines, including logistic regression, support vector machine, and k-nearest neighbors and had competitive performance relative to XGBoost. On an external independent dataset, DrugReasoner outperformed both baseline and the recently developed ChemAP model, achieving an AUC of 0.728 and an F1-score of 0.774, while maintaining high precision and balanced sensitivity, demonstrating robustness in real-world scenarios. These findings demonstrate that DrugReasoner not only delivers competitive predictive accuracy but also enhances transparency through its reasoning outputs, thereby addressing a key bottleneck in AI-assisted drug discovery. This study highlights the potential of reasoning-augmented LLMs as interpretable and effective tools for pharmaceutical decision-making.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 01:14:14 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 19:01:19 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ghaffarzadeh-Esfahani",
"Mohammadreza",
""
],
[
"Motahharynia",
"Ali",
""
],
[
"Yousefian",
"Nahid",
""
],
[
"Mazrouei",
"Navid",
""
],
[
"Ghaisari",
"Jafar",
""
],
[
"Gheisari",
"Yousof",
""
]
] |
2508.18680
|
Yen-Chi Lee
|
Yun-Feng Lo, Yen-Chi Lee
|
Joint Time-Position Statistics and Fisher Information in Drift-Diffusion Molecular Channels
|
4 pages, 4 figures
| null | null | null |
cs.IT eess.SP math.IT
|
http://creativecommons.org/licenses/by/4.0/
|
This letter derives a closed-form joint distribution of the first arrival time (FAT) and first arrival position (FAP) in diffusion-based molecular communication (MC) systems with drift. Unlike prior work that studied FAT or FAP separately, we obtain an explicit joint probability density function under constant drift and isotropic diffusion in arbitrary dimensions, revealing a nontrivial spatiotemporal coupling. Based on this result, we compute the Fisher information matrix (FIM) and show that joint observations enable estimation of lateral drift and improve sensitivity to diffusion -- capabilities not attainable with time-only or position-only models.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 04:58:53 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 05:09:51 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lo",
"Yun-Feng",
""
],
[
"Lee",
"Yen-Chi",
""
]
] |
2508.18682
|
Jingbo Liu
|
Jingbo Liu
|
Simple and Sharp Generalization Bounds via Lifting
|
74 pages, 1 figure
| null | null | null |
math.ST stat.TH
|
http://creativecommons.org/publicdomain/zero/1.0/
|
We develop an information-theoretic framework for bounding the expected supremum and tail probabilities of stochastic processes, offering a simpler and sharper alternative to classical chaining and slicing arguments for generalization bounds. The key idea is a lifting argument that produces information-theoretic analogues of empirical process bounds, such as Dudley's entropy integral. The lifting introduces symmetry, yielding sharp bounds even when the classical Dudley integral is loose. As a by-product, we obtain a concise proof of the majorizing measure theorem, providing explicit constants. The information-theoretic approach provides a soft version of classical localized complexity bounds in generalization theory, but is more concise and does not require the slicing argument. We apply this approach to empirical risk minimization over Sobolev ellipsoids and weak $\ell_q$ balls, obtaining sharper convergence rates or extensions to settings not covered by classical methods.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 05:04:54 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 02:53:20 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Liu",
"Jingbo",
""
]
] |
2508.18688
|
Hejiang Cai
|
Hejiang Cai, Di Wu, Ji Xu, Xiang Liu, Yiziting Zhu, Xin Shu, Yujie Li, Bin Yi
|
End to End Autoencoder MLP Framework for Sepsis Prediction
| null | null | null | null |
cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Sepsis is a life threatening condition that requires timely detection in intensive care settings. Traditional machine learning approaches, including Naive Bayes, Support Vector Machine (SVM), Random Forest, and XGBoost, often rely on manual feature engineering and struggle with irregular, incomplete time-series data commonly present in electronic health records. We introduce an end-to-end deep learning framework integrating an unsupervised autoencoder for automatic feature extraction with a multilayer perceptron classifier for binary sepsis risk prediction. To enhance clinical applicability, we implement a customized down sampling strategy that extracts high information density segments during training and a non-overlapping dynamic sliding window mechanism for real-time inference. Preprocessed time series data are represented as fixed dimension vectors with explicit missingness indicators, mitigating bias and noise. We validate our approach on three ICU cohorts. Our end-to-end model achieves accuracies of 74.6 percent, 80.6 percent, and 93.5 percent, respectively, consistently outperforming traditional machine learning baselines. These results demonstrate the framework's superior robustness, generalizability, and clinical utility for early sepsis detection across heterogeneous ICU environments.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 05:22:48 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 12:20:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cai",
"Hejiang",
""
],
[
"Wu",
"Di",
""
],
[
"Xu",
"Ji",
""
],
[
"Liu",
"Xiang",
""
],
[
"Zhu",
"Yiziting",
""
],
[
"Shu",
"Xin",
""
],
[
"Li",
"Yujie",
""
],
[
"Yi",
"Bin",
""
]
] |
2508.18695
|
Wasi Ullah Engr.
|
Wasi Ullah, Yasir Noman Khalid, Saddam Hussain Khan
|
A Novel Deep Hybrid Framework with Ensemble-Based Feature Optimization for Robust Real-Time Human Activity Recognition
|
35 pages, 25 figures, 11 tables
| null | null | null |
cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Human Activity Recognition (HAR) plays a pivotal role in various applications, including smart surveillance, healthcare, assistive technologies, sports analytics, etc. However, HAR systems still face critical challenges, including high computational costs, redundant features, and limited scalability in real-time scenarios. An optimized hybrid deep learning framework is introduced that integrates a customized InceptionV3, an LSTM architecture, and a novel ensemble-based feature selection strategy. The proposed framework first extracts spatial descriptors using the customized InceptionV3 model, which captures multilevel contextual patterns, region homogeneity, and fine-grained localization cues. The temporal dependencies across frames are then modeled using LSTMs to effectively encode motion dynamics. Finally, an ensemble-based genetic algorithm with Adaptive Dynamic Fitness Sharing and Attention (ADFSA) is employed to select a compact and optimized feature set by dynamically balancing objectives such as accuracy, redundancy, uniqueness, and complexity reduction. Consequently, the selected feature subsets, which are both diverse and discriminative, enable various lightweight machine learning classifiers to achieve accurate and robust HAR in heterogeneous environments. Experimental results on the robust UCF-YouTube dataset, which presents challenges such as occlusion, cluttered backgrounds, motion dynamics, and poor illumination, demonstrate good performance. The proposed approach achieves 99.65% recognition accuracy, reduces features to as few as 7, and enhances inference time. The lightweight and scalable nature of the HAR system supports real-time deployment on edge devices such as Raspberry Pi, enabling practical applications in intelligent, resource-aware environments, including public safety, assistive technology, and autonomous monitoring systems.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 05:42:55 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 09:28:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ullah",
"Wasi",
""
],
[
"Khalid",
"Yasir Noman",
""
],
[
"Khan",
"Saddam Hussain",
""
]
] |
2508.18722
|
Honghao Fu
|
Honghao Fu, Junlong Ren, Qi Chai, Deheng Ye, Yujun Cai, Hao Wang
|
VistaWise: Building Cost-Effective Agent with Cross-Modal Knowledge Graph for Minecraft
|
Accepted by EMNLP 2025 main
| null | null | null |
cs.AI
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) have shown significant promise in embodied decision-making tasks within virtual open-world environments. Nonetheless, their performance is hindered by the absence of domain-specific knowledge. Methods that finetune on large-scale domain-specific data entail prohibitive development costs. This paper introduces VistaWise, a cost-effective agent framework that integrates cross-modal domain knowledge and finetunes a dedicated object detection model for visual analysis. It reduces the requirement for domain-specific training data from millions of samples to a few hundred. VistaWise integrates visual information and textual dependencies into a cross-modal knowledge graph (KG), enabling a comprehensive and accurate understanding of multimodal environments. We also equip the agent with a retrieval-based pooling strategy to extract task-related information from the KG, and a desktop-level skill library to support direct operation of the Minecraft desktop client via mouse and keyboard inputs. Experimental results demonstrate that VistaWise achieves state-of-the-art performance across various open-world tasks, highlighting its effectiveness in reducing development costs while enhancing agent performance.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 06:39:15 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 11:01:08 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Fu",
"Honghao",
""
],
[
"Ren",
"Junlong",
""
],
[
"Chai",
"Qi",
""
],
[
"Ye",
"Deheng",
""
],
[
"Cai",
"Yujun",
""
],
[
"Wang",
"Hao",
""
]
] |
2508.18750
|
Zeng Zhang
|
Zeng Zhang and Xiaoqi Li
|
Immutable Digital Recognition via Blockchain
| null | null | null | null |
cs.CR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The development of blockchain technology has significantly enhanced the security and transparency of personal information and transaction records. Concurrent with the advancement of blockchain technology and the emergence of the digital currency ecosystem, the internet has evolved from a paradigm dominated by information flow to one driven by value flow. Consequently, the concept of token has gained widespread dissemination, and the electronic token under investigation in this thesis is a development of this concept. The application of electronic tokens has become pervasive with the development of the internet, but the functionality of these tokens is often limited, and issues related to trust remain significant challenges. This study proposes innovative solutions to address the deficiencies in traditional electronic token systems, including the issuance of tokens, the heterogeneity of issuance standards, and the lack of democracy. The solutions are based on distributed storage, anti-tampering mechanisms, consensus protocols, and transparent, traceable storage. Additionally, the study employs NFTs, smart contracts, and RSA blind signatures, among other key technologies, to construct a system based on blockchain technology. The process integrates the decentralised management and centralised operation models, aligning them with the national policy directives. The developed solution enables the full utilisation of blockchain technology's advantages while also fostering community participation. Consequently, it establishes a secure, legal, reliable, and dynamic electronic certification system.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 07:27:12 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 08:19:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Zhang",
"Zeng",
""
],
[
"Li",
"Xiaoqi",
""
]
] |
2508.18916
|
Bojan Evkoski
|
Bojan Evkoski, Igor Mozeti\v{c}, Nikola Ljube\v{s}i\'c, Petra Kralj Novak
|
Affective Polarization across European Parliaments
|
6 pages, 4 figures
| null | null | null |
cs.CL cs.SI
|
http://creativecommons.org/licenses/by/4.0/
|
Affective polarization, characterized by increased negativity and hostility towards opposing groups, has become a prominent feature of political discourse worldwide. Our study examines the presence of this type of polarization in a selection of European parliaments in a fully automated manner. Utilizing a comprehensive corpus of parliamentary speeches from the parliaments of six European countries, we employ natural language processing techniques to estimate parliamentarian sentiment. By comparing the levels of negativity conveyed in references to individuals from opposing groups versus one's own, we discover patterns of affectively polarized interactions. The findings demonstrate the existence of consistent affective polarization across all six European parliaments. Although activity correlates with negativity, there is no observed difference in affective polarization between less active and more active members of parliament. Finally, we show that reciprocity is a contributing mechanism in affective polarization between parliamentarians across all six parliaments.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 10:39:56 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 06:19:01 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Evkoski",
"Bojan",
""
],
[
"Mozetič",
"Igor",
""
],
[
"Ljubešić",
"Nikola",
""
],
[
"Novak",
"Petra Kralj",
""
]
] |
2508.18921
|
Jakub Micha\'nk\'ow
|
Jakub Micha\'nk\'ow
|
Forecasting Probability Distributions of Financial Returns with Deep Neural Networks
|
12 pages, 4 figures, 5 tables, fixed figures, repetitions
| null | null | null |
q-fin.RM cs.LG
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This study evaluates deep neural networks for forecasting probability distributions of financial returns. 1D convolutional neural networks (CNN) and Long Short-Term Memory (LSTM) architectures are used to forecast parameters of three probability distributions: Normal, Student's t, and skewed Student's t. Using custom negative log-likelihood loss functions, distribution parameters are optimized directly. The models are tested on six major equity indices (S\&P 500, BOVESPA, DAX, WIG, Nikkei 225, and KOSPI) using probabilistic evaluation metrics including Log Predictive Score (LPS), Continuous Ranked Probability Score (CRPS), and Probability Integral Transform (PIT). Results show that deep learning models provide accurate distributional forecasts and perform competitively with classical GARCH models for Value-at-Risk estimation. The LSTM with skewed Student's t distribution performs best across multiple evaluation criteria, capturing both heavy tails and asymmetry in financial returns. This work shows that deep neural networks are viable alternatives to traditional econometric models for financial risk assessment and portfolio management.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 10:48:16 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 12:37:18 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Michańków",
"Jakub",
""
]
] |
2508.19005
|
Jie Zhou
|
Yuxuan Cai, Yipeng Hao, Jie Zhou, Hang Yan, Zhikai Lei, Rui Zhen, Zhenhua Han, Yutao Yang, Junsong Li, Qianjun Pan, Tianyu Huai, Qin Chen, Xin Li, Kai Chen, Bo Zhang, Xipeng Qiu, and Liang He
|
Building Self-Evolving Agents via Experience-Driven Lifelong Learning: A Framework and Benchmark
| null | null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
As AI advances toward general intelligence, the focus is shifting from systems optimized for static tasks to creating open-ended agents that learn continuously. In this paper, we introduce Experience-driven Lifelong Learning (ELL), a framework for building self-evolving agents capable of continuous growth through real-world interaction. The framework is built on four core principles: (1) Experience Exploration: Agents learn through continuous, self-motivated interaction with dynamic environments, navigating interdependent tasks and generating rich experiential trajectories. (2) Long-term Memory: Agents preserve and structure historical knowledge, including personal experiences, domain expertise, and commonsense reasoning, into a persistent memory system. (3) Skill Learning: Agents autonomously improve by abstracting recurring patterns from experience into reusable skills, which are actively refined and validated for application in new tasks. (4) Knowledge Internalization: Agents internalize explicit and discrete experiences into implicit and intuitive capabilities as "second nature".
We also introduce StuLife, a benchmark dataset for ELL that simulates a student's holistic college journey, from enrollment to academic and personal development, across three core phases and ten detailed sub-scenarios. StuLife is designed around three key paradigm
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 13:04:28 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 03:28:14 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Cai",
"Yuxuan",
""
],
[
"Hao",
"Yipeng",
""
],
[
"Zhou",
"Jie",
""
],
[
"Yan",
"Hang",
""
],
[
"Lei",
"Zhikai",
""
],
[
"Zhen",
"Rui",
""
],
[
"Han",
"Zhenhua",
""
],
[
"Yang",
"Yutao",
""
],
[
"Li",
"Junsong",
""
],
[
"Pan",
"Qianjun",
""
],
[
"Huai",
"Tianyu",
""
],
[
"Chen",
"Qin",
""
],
[
"Li",
"Xin",
""
],
[
"Chen",
"Kai",
""
],
[
"Zhang",
"Bo",
""
],
[
"Qiu",
"Xipeng",
""
],
[
"He",
"Liang",
""
]
] |
2508.19013
|
Ioannis Hadjifrangiskou
|
Ioannis Hadjifrangiskou, Sumesh P. Thampi, Julia M. Yeomans
|
Channel flows of deformable nematics
| null | null |
10.1103/9mx1-1hk8
| null |
cond-mat.soft
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We describe channel flows in a continuum model of deformable nematic particles. In a simple shear flow, deformability leads to a nonlinear coupling of strain rate and vorticity, and results in shape oscillations or flow alignment. The final steady state can depend on initial conditions, and we explain this behaviour by considering a phase space representation of the dynamics. In Poiseuille flow, particle deformability and nematic elasticity induce banding, where particles near the walls are aligned, and those near the centre of the channel oscillate in direction and shape. Our results show that particle deformability can lead to complex behaviour even in simple flows, suggesting new microfluidic experiments.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 13:23:03 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hadjifrangiskou",
"Ioannis",
""
],
[
"Thampi",
"Sumesh P.",
""
],
[
"Yeomans",
"Julia M.",
""
]
] |
2508.19014
|
Surajit Das
|
Surajit Das, Gourav Roy, Aleksei Eliseev, and Ram Kumar Rajendran
|
MAB Optimizer for Estimating Math Question Difficulty via Inverse CV without NLP
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
The evolution of technology and education is driving the emergence of Intelligent & Autonomous Tutoring Systems (IATS), where objective and domain-agnostic methods for determining question difficulty are essential. Traditional human labeling is subjective, and existing NLP-based approaches fail in symbolic domains like algebra. This study introduces the Approach of Passive Measures among Educands (APME), a reinforcement learning-based Multi-Armed Bandit (MAB) framework that estimates difficulty solely from solver performance data -- marks obtained and time taken -- without requiring linguistic features or expert labels. By leveraging the inverse coefficient of variation as a risk-adjusted metric, the model provides an explainable and scalable mechanism for adaptive assessment. Empirical validation was conducted on three heterogeneous datasets. Across these diverse contexts, the model achieved an average R2 of 0.9213 and an average RMSE of 0.0584, confirming its robustness, accuracy, and adaptability to different educational levels and assessment formats. Compared with baseline approaches-such as regression-based, NLP-driven, and IRT models-the proposed framework consistently outperformed alternatives, particularly in purely symbolic domains. The findings highlight that (i) item heterogeneity strongly influences perceived difficulty, and (ii) variance in solver outcomes is as critical as mean performance for adaptive allocation. Pedagogically, the model aligns with Vygotskys Zone of Proximal Development by identifying tasks that balance challenge and attainability, supporting motivation while minimizing disengagement. This domain-agnostic, self-supervised approach advances difficulty tagging in IATS and can be extended beyond algebra wherever solver interaction data is available
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 13:23:31 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 21:05:21 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Das",
"Surajit",
""
],
[
"Roy",
"Gourav",
""
],
[
"Eliseev",
"Aleksei",
""
],
[
"Rajendran",
"Ram Kumar",
""
]
] |
2508.19087
|
Shaobo Ma
|
Shaobo Ma, Chao Fang, Haikuo Shao, Zhongfeng Wang
|
APT-LLM: Exploiting Arbitrary-Precision Tensor Core Computing for LLM Acceleration
|
To appear in the IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD)
| null |
10.1109/TCAD.2025.3604321
| null |
cs.LG cs.AI cs.AR
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Large language models (LLMs) have revolutionized AI applications, yet their enormous computational demands severely limit deployment and real-time performance. Quantization methods can help reduce computational costs, however, attaining the extreme efficiency associated with ultra-low-bit quantized LLMs at arbitrary precision presents challenges on GPUs. This is primarily due to the limited support for GPU Tensor Cores, inefficient memory management, and inflexible kernel optimizations. To tackle these challenges, we propose a comprehensive acceleration scheme for arbitrary precision LLMs, namely APT-LLM. Firstly, we introduce a novel data format, bipolar-INT, which allows for efficient and lossless conversion with signed INT, while also being more conducive to parallel computation. We also develop a matrix multiplication (MatMul) method allowing for arbitrary precision by dismantling and reassembling matrices at the bit level. This method provides flexible precision and optimizes the utilization of GPU Tensor Cores. In addition, we propose a memory management system focused on data recovery, which strategically employs fast shared memory to substantially increase kernel execution speed and reduce memory access latency. Finally, we develop a kernel mapping method that dynamically selects the optimal configurable hyperparameters of kernels for varying matrix sizes, enabling optimal performance across different LLM architectures and precision settings. In LLM inference, APT-LLM achieves up to a 3.99$\times$ speedup compared to FP16 baselines and a 2.16$\times$ speedup over NVIDIA CUTLASS INT4 acceleration on RTX 3090. On RTX 4090 and H800, APT-LLM achieves up to 2.44$\times$ speedup over FP16 and 1.65$\times$ speedup over CUTLASS integer baselines.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 14:48:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ma",
"Shaobo",
""
],
[
"Fang",
"Chao",
""
],
[
"Shao",
"Haikuo",
""
],
[
"Wang",
"Zhongfeng",
""
]
] |
2508.19156
|
Felix Hahne
|
Felix Hahne, Teresa Klara Pfau, Liza \v{Z}aper, Lucio Stefan, Thibault Capelle, Andrea Ranfagni, Martino Poggio, Albert Schliesser
|
Measuring high field gradients of cobalt nanomagnets in a spin-mechanical setup
| null | null | null | null |
cond-mat.mes-hall quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
Hybrid systems composed of a single nitrogen-vacancy center spin magnetically coupled to a macroscopic mechanical resonator constitute promising platforms for the realization of quantum information protocols and for quantum sensing applications. The magnetic structure that mediates the interaction must ensure high field gradients while preserving the spin and mechanical properties. We present a spin-mechanical setup built around a cobalt nanomagnet grown with focused electron beam-induced deposition. The magnetic structure is fully characterized, and a maximum gradient of $170\,\mathrm{kT/m}$ is directly measured at a spin-oscillator distance of a few hundred nanometers. Spin coherence was preserved at the value of $20\,\mathrm{ \mu s}$ up to a gradient of $25\,\mathrm{kT/m}$. The effect of the mechanical motion onto the spin dynamics was observed, thus signifying the presence of spin-mechanics coupling. Given the noninvasive nature of the nanomagnet deposition process, we foresee the adoption of such structures in hybrid platforms with high-quality factor resonators, in the "magnet on oscillator" configuration.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 16:08:01 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:37:26 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hahne",
"Felix",
""
],
[
"Pfau",
"Teresa Klara",
""
],
[
"Žaper",
"Liza",
""
],
[
"Stefan",
"Lucio",
""
],
[
"Capelle",
"Thibault",
""
],
[
"Ranfagni",
"Andrea",
""
],
[
"Poggio",
"Martino",
""
],
[
"Schliesser",
"Albert",
""
]
] |
2508.19231
|
Raihaneh Moti
|
Raihaneh Moti and Ali Shojai
|
On the Dynamics of Extended Bodies and the GravoThermo Memory Effect
|
28 pages, 13 figures
| null | null | null |
gr-qc hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The non-linearity of the theory of gravity induces a hysteresis effect in both the systems interacting with gravity and in the gravitational field. The effect is usually referred to as the memory effect. In this paper, we explore this phenomenon in the context of the dynamics of extended rotating objects in the presence of a gravitational wave pulse. Then, we consider an ensemble of such objects and show that a redistribution of spin orientations takes place as a gravitational wave passes through. We examine how this redistribution, changes the partition function and the thermodynamic quantities including entropy and energy of the system. The result shows that some information about the source of the gravitation wave is encoded in the thermodynamics of the system.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 17:50:33 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 08:17:58 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Moti",
"Raihaneh",
""
],
[
"Shojai",
"Ali",
""
]
] |
2508.19302
|
Avery Carr
|
Avery Carr
|
Cycles of Length 4 or 8 in Graphs with Diameter 2 and Minimum Degree at Least 3
|
4 pages, 4 figures. Short note. Relates to the Erd\H{o}s--Gy\'arf\'as Conjecture
| null | null | null |
math.CO
|
http://creativecommons.org/licenses/by/4.0/
|
In this short note it is shown that every graph of diameter 2 and minimum degree at least 3 contains a cycle of length 4 or 8. This result contributes to the study of the Erd\H{o}s--Gy\'arf\'as Conjecture.
|
[
{
"version": "v1",
"created": "Mon, 25 Aug 2025 21:27:09 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 18:42:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Carr",
"Avery",
""
]
] |
2508.19340
|
Ivan Minchev
|
J.P. Bernaldez, I. Minchev, B. Ratcliffe, L. Marques, K. Sysoliatina, J. Walcher, S. Khoperskov, M. Martig, R. de Jong, M. Steinmetz
|
The Impact of Radial Migration on Disk Galaxy SFHs: II. The Role of bar strength, disk thickness, and merger history
|
19 pages, 15 figures, Submitted to A&A. Comments welcome!
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Reconstructing the star formation history (SFH) of disk galaxies is central to understanding their growth and evolution, yet such estimates can be strongly biased by stellar radial migration - the redistribution of angular momenta over cosmic time. Using 186 Milky Way (MW) and Andromeda (M31) analogues from the TNG50 cosmological simulation, we compare star formation rates (SFRs) inferred from present-day stellar positions with those based on stellar birth radii to quantify the magnitude, spatial structure, and temporal evolution of migration-induced biases. We find that radial migration systematically produces artificial star formation in regions that had not yet formed stars. Notably, about 80% of galaxies exhibit outer-disk stars older than 10 Gyr, implying that these populations must have migrated from the inner disk. Similar effects appear in about 45% of galaxies at intermediate radii during early epochs, and in 30% of quenched inner disks within the past 4 Gyr. Migration also smooths SFHs, washing out localized bursts and periods of suppressed star formation by dispersing stars across neighboring radii. The strength and imprint of these distortions depend sensitively on galactic structure and evolutionary history: strong bars drive mean SFR overestimates of up to 75% in the inner disk and 150% in the outskirts; thinner, dynamically cold disks suffer average outer-disk biases up to 160%; while thick disks exhibit typical inner-disk biases up to 125%. Merger timing further modulates these patterns. Our results demonstrate that failing to account for stellar migration can lead to severe misinterpretations of when and where stars formed, with direct implications for the chemical and evolutionary histories of the MW and external galaxies.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 18:00:20 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 14:40:39 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bernaldez",
"J. P.",
""
],
[
"Minchev",
"I.",
""
],
[
"Ratcliffe",
"B.",
""
],
[
"Marques",
"L.",
""
],
[
"Sysoliatina",
"K.",
""
],
[
"Walcher",
"J.",
""
],
[
"Khoperskov",
"S.",
""
],
[
"Martig",
"M.",
""
],
[
"de Jong",
"R.",
""
],
[
"Steinmetz",
"M.",
""
]
] |
2508.19448
|
Ludovico Theo Giorgini
|
Ludovico Theo Giorgini, Tobias Bischoff, Andre Noguiera Souza
|
Reduced-Order Modeling of Cyclo-Stationary Time Series Using Score-Based Generative Methods
| null | null | null | null |
nlin.CD stat.ML
|
http://creativecommons.org/licenses/by/4.0/
|
Many natural systems exhibit cyclo-stationary behavior characterized by periodic forcing such as annual and diurnal cycles. We present a data-driven method leveraging recent advances in score-based generative modeling to construct reduced-order models for such cyclo-stationary time series. Our approach accurately reproduces the statistical properties and temporal correlations of the original data, enabling efficient generation of synthetic trajectories. We demonstrate the performance of the method through application to the Planet Simulator (PlaSim) climate model, constructing a reduced-order model for the 20 leading principal components of surface temperature driven by the annual cycle. The resulting surrogate model accurately reproduces the marginal and joint probability distributions, autocorrelation functions, and spatial coherence of the original climate system across multiple validation metrics. The approach offers substantial computational advantages, enabling generation of centuries of synthetic climate data in minutes compared to weeks required for equivalent full model simulations. This work opens new possibilities for efficient modeling of periodically forced systems across diverse scientific domains, providing a principled framework for balancing computational efficiency with physical fidelity in reduced-order modeling applications.
|
[
{
"version": "v1",
"created": "Tue, 26 Aug 2025 21:49:48 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 14:45:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Giorgini",
"Ludovico Theo",
""
],
[
"Bischoff",
"Tobias",
""
],
[
"Souza",
"Andre Noguiera",
""
]
] |
2508.19501
|
Satoshi Takada
|
Ryu Suzuki, Shintaro Hokada, Satoshi Takada
|
Stress Analysis of a Square Elastic Body Under Biaxial Loading Using Airy Stress Functions
|
19 pages, 8 figures
| null | null | null |
physics.class-ph cond-mat.soft
|
http://creativecommons.org/licenses/by/4.0/
|
This study presents an analytical investigation of stress distributions in square-shaped elastic bodies subjected to concentrated compressive loads under uniaxial and biaxial conditions. By employing the Airy stress function method, we derive closed-form solutions that satisfy the governing biharmonic equation and the prescribed boundary conditions along the edges of the square domain. The stress components are expressed as series expansions, with coefficients determined to enforce boundary constraints. In the uniaxial compression case, the resulting stress fields exhibit strong agreement with photoelastic fringe patterns previously observed in experimental studies. For biaxial loading, the solution represents a superposition of two orthogonal compression scenarios, producing spatial variations in the principal stress difference depending on the location within the domain.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 01:12:49 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Suzuki",
"Ryu",
""
],
[
"Hokada",
"Shintaro",
""
],
[
"Takada",
"Satoshi",
""
]
] |
2508.19524
|
Rahim Moosa
|
Rahim Moosa and Anand Pillay
|
Definable Galois theory for bimeromorphic geometry
|
33 pages
| null | null | null |
math.LO math.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The outlines of a "Galois theory" for bimeromorphic geometry is here developed, via the study of model-theoretic definable binding groups in the theory CCM of compact complex spaces. As an application, a structure theorem about principal meromorphic bundles with algebraic structure group, and admitting no horizontal subvarieties, is deduced. Examples of algebraic groups arising as binding groups are provided, as is a characterisation of when they are linear. Using binding groups in CCM it is shown that, in contrast to the situation in differentially closed fields, there are many algebraic groups which admit nontrivial definable torsors over acl-closed sets in the theory DCCM of existentially closed differential CCM-structures. A self-contained exposition of the binding group theorem in totally transcendental theories, that emphasises the bitorsorial nature of the construction, is also included.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 02:36:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Moosa",
"Rahim",
""
],
[
"Pillay",
"Anand",
""
]
] |
2508.19525
|
Tianshi Xu
|
Tianshi Xu, Wen-jie Lu, Jiangrui Yu, Chen Yi, Chenqi Lin, Runsheng Wang, Meng Li
|
Breaking the Layer Barrier: Remodeling Private Transformer Inference with Hybrid CKKS and MPC
|
USENIX Security 2025
| null | null | null |
cs.CR
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
This paper presents an efficient framework for private Transformer inference that combines Homomorphic Encryption (HE) and Secure Multi-party Computation (MPC) to protect data privacy. Existing methods often leverage HE for linear layers (e.g., matrix multiplications) and MPC for non-linear layers (e.g., Softmax activation functions), but the conversion between HE and MPC introduces significant communication costs. The proposed framework, dubbed BLB, overcomes this by breaking down layers into fine-grained operators and further fusing adjacent linear operators, reducing the need for HE/MPC conversions. To manage the increased ciphertext bit width from the fused linear operators, BLB proposes the first secure conversion protocol between CKKS and MPC and enables CKKS-based computation of the fused operators. Additionally, BLB proposes an efficient matrix multiplication protocol for fused computation in Transformers. Extensive evaluations on BERT-base, BERT-large, and GPT2-base show that BLB achieves a $21\times$ reduction in communication overhead compared to BOLT (S\&P'24) and a $2\times$ reduction compared to Bumblebee (NDSS'25), along with latency reductions of $13\times$ and $1.8\times$, respectively, when leveraging GPU acceleration.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 02:40:50 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 13:24:24 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Xu",
"Tianshi",
""
],
[
"Lu",
"Wen-jie",
""
],
[
"Yu",
"Jiangrui",
""
],
[
"Yi",
"Chen",
""
],
[
"Lin",
"Chenqi",
""
],
[
"Wang",
"Runsheng",
""
],
[
"Li",
"Meng",
""
]
] |
2508.19549
|
Sukdev Mahapatra
|
Sukdev Mahapatra
|
Modeling Orbital Decay of Low-Earth Orbit Satellites due to Atmospheric Drag: A Simplified Analytical Approach
|
7 pages, 4 figures
| null | null | null |
astro-ph.EP astro-ph.IM
|
http://creativecommons.org/licenses/by/4.0/
|
Low-Earth Orbit (LEO) satellites are crucial for communications, navigation, and Earth observation. However, their operational lifetimes are strongly influenced by orbital decay due to atmospheric drag. This work presents a simplified analytical model of orbital decay in LEO, incorporating atmospheric density variations and satellite parameters. Using standard drag force equations and density models, we explore how altitude, satellite mass, and cross-sectional area affect decay timescales. Comparisons are made with existing NASA models and previous literature. The results highlight the importance of accurate atmospheric density representation and solar activity in predicting satellite lifetimes, especially relevant in the context of increasing space debris and mega-constellations.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 03:50:22 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 05:04:43 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mahapatra",
"Sukdev",
""
]
] |
2508.19611
|
Huaiyuan Yao
|
Huaiyuan Yao, Wanpeng Xu, Justin Turnau, Nadia Kellam, Hua Wei
|
Instructional Agents: LLM Agents on Automated Course Material Generation for Teaching Faculties
|
18 pages, 9 figures
| null | null | null |
cs.AI cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Preparing high-quality instructional materials remains a labor-intensive process that often requires extensive coordination among teaching faculty, instructional designers, and teaching assistants. In this work, we present Instructional Agents, a multi-agent large language model (LLM) framework designed to automate end-to-end course material generation, including syllabus creation, lecture scripts, LaTeX-based slides, and assessments. Unlike existing AI-assisted educational tools that focus on isolated tasks, Instructional Agents simulates role-based collaboration among educational agents to produce cohesive and pedagogically aligned content. The system operates in four modes: Autonomous, Catalog-Guided, Feedback-Guided, and Full Co-Pilot mode, enabling flexible control over the degree of human involvement. We evaluate Instructional Agents across five university-level computer science courses and show that it produces high-quality instructional materials while significantly reducing development time and human workload. By supporting institutions with limited instructional design capacity, Instructional Agents provides a scalable and cost-effective framework to democratize access to high-quality education, particularly in underserved or resource-constrained settings.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 06:45:06 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 01:38:20 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yao",
"Huaiyuan",
""
],
[
"Xu",
"Wanpeng",
""
],
[
"Turnau",
"Justin",
""
],
[
"Kellam",
"Nadia",
""
],
[
"Wei",
"Hua",
""
]
] |
2508.19678
|
Chao Wang
|
Chao Wang, Shuyuan Zhang, Lei Wang
|
Distributed Safety-Critical MPC for Multi-Agent Formation Control and Obstacle Avoidance
|
Accepted for presentation at the 64th IEEE Conference on Decision and Control (CDC 2025)
| null | null | null |
eess.SY cs.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
For nonlinear multi-agent systems with high relative degrees, achieving formation control and obstacle avoidance in a distributed manner remains a significant challenge. To address this issue, we propose a novel distributed safety-critical model predictive control (DSMPC) algorithm that incorporates discrete-time high-order control barrier functions (DHCBFs) to enforce safety constraints, alongside discrete-time control Lyapunov functions (DCLFs) to establish terminal constraints. To facilitate distributed implementation, we develop estimated neighbor states for formulating DHCBFs and DCLFs, while also devising a bound constraint to limit estimation errors and ensure convergence. Additionally, we provide theoretical guarantees regarding the feasibility and stability of the proposed DSMPC algorithm based on a mild assumption. The effectiveness of the proposed method is evidenced by the simulation results, demonstrating improved performance and reduced computation time compared to existing approaches.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 08:39:35 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 07:59:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Chao",
""
],
[
"Zhang",
"Shuyuan",
""
],
[
"Wang",
"Lei",
""
]
] |
2508.19681
|
Jun Xu
|
Hai-Cheng Wang, Song-Jie Li, Jun Xu, and Zhong-Zhou Ren
|
A comparison study of collisions at relativistic energies involving light nuclei
|
8 pages, 6 figures
|
Universe 11, 296 (2025)
|
10.3390/universe11090296
| null |
nucl-th hep-ex nucl-ex
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
We present extensive comparisons of $^{16}$O+$^{16}$O collisions at the center-of-mass energy per nucleon pair $\sqrt{s_{NN}}=200$ GeV and $^{208}$Pb+$^{16}$O collisions at $\sqrt{s_{NN}}=68.5$ GeV as well as $^{20}$Ne+$^{20}$Ne collisions at $\sqrt{s_{NN}}=200$ GeV and $^{208}$Pb+$^{20}$Ne collisions at $\sqrt{s_{NN}}=68.5$ GeV based on a multiphase transport (AMPT) model. We recommend measuring the ratio of the elliptic flow to the triangular flow, which shows appreciable sensitivity to the structure of light nuclei as also found in other studies. This is especially so if the observable is measured near the target rapidity in $^{208}$Pb+$^{16}$O or $^{208}$Pb+$^{20}$Ne collisions, as originally found in the present study. Our study serves as a useful reference for understanding the structure effect on observables in collisions involving light nuclei under analysis or on the schedule.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 08:41:09 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 06:45:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Hai-Cheng",
""
],
[
"Li",
"Song-Jie",
""
],
[
"Xu",
"Jun",
""
],
[
"Ren",
"Zhong-Zhou",
""
]
] |
2508.19701
|
\'Agnes K\'osp\'al
|
\'A. K\'osp\'al, P. \'Abrah\'am, V. V. Akimkin, L. Chen, J. Forbrich, K. V. Getman, B. Portilla-Revelo, D. Semenov, S. E. van Terwisga, J. Varga, L. Zwicky, G. G. Bal\'azs, Zs. Bora, \'A. Horti-D\'avid, A. P. Jo\'o, W. Og{\l}oza, B. Seli, M. Siwak, \'A. S\'odor, N. Tak\'acs
|
Time-resolved protoplanetary disk physics in DQ Tau with JWST
|
21 pages, 20 figures, accepted for publication in the A&A
| null | null | null |
astro-ph.SR astro-ph.EP
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Accretion variability is ubiquitous in YSOs. While large outbursts may strongly affect the disk, the effects of moderate bursts are less understood. We aim to study the physical response of the disk around the eccentric binary system DQ Tau to its periodic accretion changes. We organized a multi-wavelength campaign centered on four JWST/MIRI spectra. We targeted three periastrons (high accretion state) and one apastron (quiescence). We used optical and near-infrared spectroscopy and photometry to measure how the accretion luminosity varies. We decomposed the multi-epoch SEDs into stellar, accretion, and rim components. We fitted the solid-state features using various opacity curves and the molecular features using slab models. We find the inner disk of DQ Tau to be highly dynamic. The temperature, luminosity, and location of the inner dust rim vary in response to the movement of stars and the L_acc variations. This causes variable shadowing of the outer disk, leading to an anti-correlation between the rim temperature and the strength of the silicate feature. The dust mineralogy remains constant, dominated by large amorphous olivine and pyroxene grains, with smaller fractions of crystalline forsterite. The excitation of CO (1550-2260 K), HCN (880-980 K), and hot H2O (740-860 K) molecules as well as the luminosity of the [NeII] line correlate with the accretion rate, while the warm (650 K) and cold (170-200 K) H2O components are mostly constant. CO emission, originating from a hot (>1500 K) region likely within the dust sublimation radius, is most sensitive to L_acc changes. In comparison with other T Tauri disks, DQ Tau is highly C-poor and displays moderately inefficient pebble drift. We conclude that even moderate accretion rate changes affect the thermal structure in the planet-forming disk regions on short timescales, providing a crucial benchmark for understanding disk evolution.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 09:08:03 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 13:08:55 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kóspál",
"Á.",
""
],
[
"Ábrahám",
"P.",
""
],
[
"Akimkin",
"V. V.",
""
],
[
"Chen",
"L.",
""
],
[
"Forbrich",
"J.",
""
],
[
"Getman",
"K. V.",
""
],
[
"Portilla-Revelo",
"B.",
""
],
[
"Semenov",
"D.",
""
],
[
"van Terwisga",
"S. E.",
""
],
[
"Varga",
"J.",
""
],
[
"Zwicky",
"L.",
""
],
[
"Balázs",
"G. G.",
""
],
[
"Bora",
"Zs.",
""
],
[
"Horti-Dávid",
"Á.",
""
],
[
"Joó",
"A. P.",
""
],
[
"Ogłoza",
"W.",
""
],
[
"Seli",
"B.",
""
],
[
"Siwak",
"M.",
""
],
[
"Sódor",
"Á.",
""
],
[
"Takács",
"N.",
""
]
] |
2508.19720
|
Yilin Wang
|
Yilin Wang and Heng Wang and Yuyang Bai and Minnan Luo
|
Continuously Steering LLMs Sensitivity to Contextual Knowledge with Proxy Models
|
emnlp 2025
| null | null | null |
cs.CL
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In Large Language Models (LLMs) generation, there exist knowledge conflicts and scenarios where parametric knowledge contradicts knowledge provided in the context. Previous works studied tuning, decoding algorithms, or locating and editing context-aware neurons to adapt LLMs to be faithful to new contextual knowledge. However, they are usually inefficient or ineffective for large models, not workable for black-box models, or unable to continuously adjust LLMs' sensitivity to the knowledge provided in the context. To mitigate these problems, we propose CSKS (Continuously Steering Knowledge Sensitivity), a simple framework that can steer LLMs' sensitivity to contextual knowledge continuously at a lightweight cost. Specifically, we tune two small LMs (i.e. proxy models) and use the difference in their output distributions to shift the original distribution of an LLM without modifying the LLM weights. In the evaluation process, we not only design synthetic data and fine-grained metrics to measure models' sensitivity to contextual knowledge but also use a real conflict dataset to validate CSKS's practical efficacy. Extensive experiments demonstrate that our framework achieves continuous and precise control over LLMs' sensitivity to contextual knowledge, enabling both increased sensitivity and reduced sensitivity, thereby allowing LLMs to prioritize either contextual or parametric knowledge as needed flexibly. Our data and code are available at https://github.com/OliveJuiceLin/CSKS.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 09:30:24 GMT"
},
{
"version": "v2",
"created": "Thu, 28 Aug 2025 10:00:55 GMT"
},
{
"version": "v3",
"created": "Sat, 30 Aug 2025 05:54:41 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Wang",
"Yilin",
""
],
[
"Wang",
"Heng",
""
],
[
"Bai",
"Yuyang",
""
],
[
"Luo",
"Minnan",
""
]
] |
2508.19751
|
Mitchell Cox Prof
|
Joshua R. Jandrell and Mitchell A. Cox
|
High-Fidelity Prediction of Perturbed Optical Fields using Fourier Feature Networks
| null | null | null | null |
physics.optics cs.LG physics.comp-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Predicting the effects of physical perturbations on optical channels is critical for advanced photonic devices, but existing modelling techniques are often computationally intensive or require exhaustive characterisation. We present a novel data-efficient machine learning framework that learns the perturbation-dependent transmission matrix of a multimode fibre. To overcome the challenge of modelling the resulting highly oscillatory functions, we encode the perturbation into a Fourier Feature basis, enabling a compact multi-layer perceptron to learn the mapping with high fidelity. On experimental data from a compressed fibre, our model predicts the output field with a 0.995 complex correlation to the ground truth, improving accuracy by an order of magnitude over standard networks while using 85\% fewer parameters. This approach provides a general tool for modelling complex optical systems from sparse measurements.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 10:25:57 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 09:02:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jandrell",
"Joshua R.",
""
],
[
"Cox",
"Mitchell A.",
""
]
] |
2508.19767
|
Karsten Brogaard
|
K. Brogaard, A. Miglio, T. Arentoft, J. S. Thomsen, G. Casali, L. Martinelli, E. Willett, and M. Tailo
|
Attempting an accurate age estimate of the open cluster NGC 6633 using CoRoT and Gaia
|
12 pages, 5 figures, accepted for publication in A&A, layout & affiliation corrections relative to v1
| null | null | null |
astro-ph.SR astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
Asteroseismology of solar-like oscillations in giant stars allow the derivation of their masses and radii. For members of open clusters, this provides an age of the cluster that should be identical to the one derived from the colour-magnitude diagram, but independent of the uncertainties that are present for that type of analysis. We aim to identify and measure the properties of giant members of the open cluster NGC6633, and combine these with asteroseismic measurements to derive a precise and self-consistent cluster age. Importantly, we wish to constrain the effects of rotational mixing on stellar evolution, since assumptions about internal mixing can have a significant impact on stellar age estimates. We identify five giant members of NGC6633 using photometry and Gaia data, supplemented by spectroscopic literature measurements. These are combined with asteroseismic measurements from CoRoT data and compared to stellar-model isochrones. This constrains the interior mixing to a low level and enables the most precise, accurate and self-consistent age estimate so far for this cluster. Asteroseismology of the giants and the cluster colour-magnitude diagram provide self-consistent masses of the giant members and their radii constrain the stellar interior mixing to a low level. The [C/N] ratios and Li abundances also suggest that rotation has had very little influence on the evolution of the stars in NGC 6633. This results in an age estimate of 0.55+0.05-0.10 Gyr for NGC6633. Four giant members appear to be in the helium-core burning evolutionary phase as also expected from evolutionary timescales. For the largest giant, the evidence remains inconclusive. A comparison to other age and mass estimates for the same stars in the literature uncovers biases for automated age estimates of helium-core burning stars.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 10:46:00 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 09:31:13 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Brogaard",
"K.",
""
],
[
"Miglio",
"A.",
""
],
[
"Arentoft",
"T.",
""
],
[
"Thomsen",
"J. S.",
""
],
[
"Casali",
"G.",
""
],
[
"Martinelli",
"L.",
""
],
[
"Willett",
"E.",
""
],
[
"Tailo",
"M.",
""
]
] |
2508.19793
|
Hristo Tonchev
|
Hristo Tonchev, Rosen Bahtev
|
Grover's search with an oracle distinguishing between solutions
|
23 pages, 14 figures
| null | null | null |
quant-ph
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Here we suggest a modification of Grover's algorithm, based on a multiphase oracle which marks each solution with a different phase when there is more than one solution. Such a modification can be used to maintain a high probability of finding a solution for a number of iterations equal to or more than the one required by the deterministic Grover's algorithm (the one based on generalized Householder reflections). We use various semiempirical methods to show that the interval of number of iterations for which the algorithm keeps the probability of finding solution high depends on the register size and the oracle phases.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 11:19:07 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 10:16:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Tonchev",
"Hristo",
""
],
[
"Bahtev",
"Rosen",
""
]
] |
2508.19933
|
Sten Elling Tingstad Jacobsen
|
Sten Elling Tingstad Jacobsen, Bal\'azs Kulcs\'ar, Anders Lindman
|
Combined Stochastic and Robust Optimization for Electric Autonomous Mobility-on-Demand with Nested Benders Decomposition
|
29 pages, 12 figures
| null | null | null |
eess.SY cs.SY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
The electrification and automation of mobility are reshaping how cities operate on-demand transport systems. Managing Electric Autonomous Mobility-on-Demand (EAMoD) fleets effectively requires coordinating dispatch, rebalancing, and charging decisions under multiple uncertainties, including travel demand, travel time, energy consumption, and charger availability. We address this challenge with a combined stochastic and robust model predictive control (MPC) framework. The framework integrates spatio-temporal Bayesian neural network forecasts with a multi-stage stochastic optimization model, formulated as a large-scale mixed-integer linear program. To ensure real-time applicability, we develop a tailored Nested Benders Decomposition that exploits the scenario tree structure and enables efficient parallelized solution. Stochastic optimization is employed to anticipate demand and infrastructure variability, while robust constraints on energy consumption and travel times safeguard feasibility under worst-case realizations. We evaluate the framework using high-fidelity simulations of San Francisco and Chicago. Compared with deterministic, reactive, and robust baselines, the combined stochastic and robust approach reduces median passenger waiting times by up to 36% and 95th-percentile delays by nearly 20%, while also lowering rebalancing distance by 27% and electricity costs by more than 35%. We also conduct a sensitivity analysis of battery size and vehicle efficiency, finding that energy-efficient vehicles maintain stable performance even with small batteries, whereas less efficient vehicles require larger batteries and greater infrastructure support. Our results emphasize the importance of jointly optimizing predictive control, vehicle capabilities, and infrastructure planning to enable scalable, cost-efficient EAMoD operations.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 14:48:34 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 09:22:39 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Jacobsen",
"Sten Elling Tingstad",
""
],
[
"Kulcsár",
"Balázs",
""
],
[
"Lindman",
"Anders",
""
]
] |
2508.19944
|
Taebaek Hwang
|
Taebaek Hwang, Minseo Kim, Gisang Lee, Seonuk Kim, Hyunjun Eun
|
KRETA: A Benchmark for Korean Reading and Reasoning in Text-Rich VQA Attuned to Diverse Visual Contexts
|
Accepted to EMNLP 2025 (Main Conference)
| null | null | null |
cs.CV cs.CL
|
http://creativecommons.org/licenses/by/4.0/
|
Understanding and reasoning over text within visual contexts poses a significant challenge for Vision-Language Models (VLMs), given the complexity and diversity of real-world scenarios. To address this challenge, text-rich Visual Question Answering (VQA) datasets and benchmarks have emerged for high-resource languages like English. However, a critical gap persists for low-resource languages such as Korean, where the lack of comprehensive benchmarks hinders robust model evaluation and comparison. To bridge this gap, we introduce KRETA, a benchmark for Korean Reading and rEasoning in Text-rich VQA Attuned to diverse visual contexts. KRETA facilitates an in-depth evaluation of both visual text understanding and reasoning capabilities, while also supporting a multifaceted assessment across 15 domains and 26 image types. Additionally, we introduce a semi-automated VQA generation pipeline specifically optimized for text-rich settings, leveraging refined stepwise image decomposition and a rigorous seven-metric evaluation protocol to ensure data quality. While KRETA is tailored for Korean, we hope our adaptable and extensible pipeline will facilitate the development of similar benchmarks in other languages, thereby accelerating multilingual VLM research. The code and dataset for KRETA are available at https://github.com/tabtoyou/KRETA.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 15:01:02 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 10:33:09 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hwang",
"Taebaek",
""
],
[
"Kim",
"Minseo",
""
],
[
"Lee",
"Gisang",
""
],
[
"Kim",
"Seonuk",
""
],
[
"Eun",
"Hyunjun",
""
]
] |
2508.20050
|
Sindre Brattegard
|
Sindre Brattegard, Thom\'as Fogarty, Thomas Busch, and Mark T. Mitchison
|
Correlated decoherence and thermometry with mobile impurities in a 1D Fermi gas
|
V2: Added references. 15 pages, 11 figures, comments are welcome
| null | null | null |
cond-mat.quant-gas cond-mat.stat-mech quant-ph
|
http://creativecommons.org/licenses/by/4.0/
|
We theoretically investigate the correlated decoherence dynamics of two mobile impurities trapped within a gas of ultracold fermionic atoms. We use a mean-field approximation to self-consistently describe the effect of impurity-gas collisions on impurity motion, while decoherence of the impurities' internal state is computed exactly within a functional determinant approach. At equilibrium, we find that the impurities undergo bath-induced localisation as the impurity-gas interaction strength is increased. We then study the non-equilibrium dynamics induced by a sudden change of the impurities' internal state, which can be experimentally probed by Ramsey interferometry. Our theoretical approach allows us to investigate the effect of impurity motion on decoherence dynamics, finding strong deviations from the universal behaviour associated with Anderson's orthogonality catastrophe when the mass imbalance between impurity and gas atoms is small. Finally, we show that mobile impurities can be used as thermometers of their environment and that bath-mediated correlations can be beneficial for thermometric performance at low temperatures, even in the presence of non-trivial impurity motion. Our results showcase the interesting open quantum dynamics of mobile impurities dephasing in a common environment, and could help provide more precise temperature estimates of ultracold fermionic mixtures.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 17:01:38 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 10:13:25 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Brattegard",
"Sindre",
""
],
[
"Fogarty",
"Thomás",
""
],
[
"Busch",
"Thomas",
""
],
[
"Mitchison",
"Mark T.",
""
]
] |
2508.20077
|
Milena Radenkovic
|
Tao Xiuyuan, Milena Radenkovic
|
ML-MaxProp: Bridging Machine Learning and Delay-Tolerant Routing for Resilient Post-Disaster Communication
| null | null | null | null |
cs.NI
|
http://creativecommons.org/licenses/by/4.0/
|
In disaster-stricken and large-scale urban emergency scenarios, ensuring reliable communication remains a formidable challenge, as collapsed infrastructure, unpredictable mobility, and severely constrained resources disrupt conventional networks. Delay-Tolerant Networks (DTNs), though resilient through their store-carry-forward paradigm, reveal the fundamental weaknesses of classical protocols - Epidemic, Spray-and-Wait, and MaxProp - when confronted with sparse encounters, buffer shortages, and volatile connectivity. To address these obstacles, this study proposes ML-MaxProp, a hybrid routing protocol that strengthens MaxProp with supervised machine learning. By leveraging contextual features such as encounter frequency, hop count, buffer occupancy, message age, and time-to-live (TTL), ML-MaxProp predicts relay suitability in real time, transforming rigid heuristics into adaptive intelligence. Extensive simulations in the ONE environment using the Helsinki SPMBM mobility model show that ML-MaxProp consistently surpasses baseline protocols, achieving higher delivery probability, lower latency, and reduced overhead. Statistical validation further shows that these improvements are both significant and robust, even under highly resource-constrained and unstable conditions. Overall, this work shows that ML-MaxProp is not just an incremental refinement but a lightweight, adaptive, and practical solution to one of the hardest challenges in DTNs: sustaining mission-critical communication when infrastructure collapses and every forwarding decision becomes critical.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 17:43:34 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 14:40:44 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Xiuyuan",
"Tao",
""
],
[
"Radenkovic",
"Milena",
""
]
] |
2508.20085
|
Zhecheng Yuan
|
Zhecheng Yuan, Tianming Wei, Langzhe Gu, Pu Hua, Tianhai Liang, Yuanpei Chen, Huazhe Xu
|
HERMES: Human-to-Robot Embodied Learning from Multi-Source Motion Data for Mobile Dexterous Manipulation
| null | null | null | null |
cs.RO
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Leveraging human motion data to impart robots with versatile manipulation skills has emerged as a promising paradigm in robotic manipulation. Nevertheless, translating multi-source human hand motions into feasible robot behaviors remains challenging, particularly for robots equipped with multi-fingered dexterous hands characterized by complex, high-dimensional action spaces. Moreover, existing approaches often struggle to produce policies capable of adapting to diverse environmental conditions. In this paper, we introduce HERMES, a human-to-robot learning framework for mobile bimanual dexterous manipulation. First, HERMES formulates a unified reinforcement learning approach capable of seamlessly transforming heterogeneous human hand motions from multiple sources into physically plausible robotic behaviors. Subsequently, to mitigate the sim2real gap, we devise an end-to-end, depth image-based sim2real transfer method for improved generalization to real-world scenarios. Furthermore, to enable autonomous operation in varied and unstructured environments, we augment the navigation foundation model with a closed-loop Perspective-n-Point (PnP) localization mechanism, ensuring precise alignment of visual goals and effectively bridging autonomous navigation and dexterous manipulation. Extensive experimental results demonstrate that HERMES consistently exhibits generalizable behaviors across diverse, in-the-wild scenarios, successfully performing numerous complex mobile bimanual dexterous manipulation tasks. Project Page:https://gemcollector.github.io/HERMES/.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 17:53:46 GMT"
},
{
"version": "v2",
"created": "Thu, 28 Aug 2025 03:19:49 GMT"
},
{
"version": "v3",
"created": "Sun, 31 Aug 2025 06:47:56 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yuan",
"Zhecheng",
""
],
[
"Wei",
"Tianming",
""
],
[
"Gu",
"Langzhe",
""
],
[
"Hua",
"Pu",
""
],
[
"Liang",
"Tianhai",
""
],
[
"Chen",
"Yuanpei",
""
],
[
"Xu",
"Huazhe",
""
]
] |
2508.20142
|
Dinesh Visva Gunasekeran
|
Samantha Bei Yi Yan, Dinesh Visva Gunasekeran, Caitlyn Tan, Kai En Chan, Caleb Tan, Charmaine Shi Min Lim, Audrey Chia, Hsien-Hsien Lei, Robert Morris, Janice Huiqin Weng (for the MINDLINE Study Group)
|
Evaluation of A National Digitally-Enabled Health Promotion Campaign for Mental Health Awareness using Social Media Platforms Tik Tok, Facebook, Instagram, and YouTube
| null | null | null | null |
cs.SI cs.CY
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Mental health disorders rank among the 10 leading contributors to the global burden of diseases, yet persistent stigma and care barriers delay early intervention. This has inspired efforts to leverage digital platforms for scalable health promotion to engage at-risk populations. To evaluate the effectiveness of a digitally-enabled mental health promotion (DEHP) campaign, we conducted an observational cross-sectional study of a 3-month (February-April 2025) nation-wide campaign in Singapore. Campaign materials were developed using a marketing funnel framework and disseminated across YouTube, Facebook, Instagram, and TikTok. This included narrative videos and infographics to promote symptom awareness, coping strategies, and/or patient navigation to Singapore's Mindline website, as the intended endpoint for user engagement and support. Primary outcomes include anonymised performance analytics (impressions, unique reach, video content view, engagements) stratified by demographics, device types, and sector. Secondary outcomes measured cost-efficiency metrics and traffic to the Mindline website respectively. This campaign generated 3.49 million total impressions and reached 1.39 million unique residents, with a Cost per Mille at $26.90, Cost per Click at $29.33, and Cost per Action at $6.06. Narrative videos accumulated over 630,000 views and 18,768 engagements. Overall, we demonstrate that DEHP campaigns can achieve national engagement for mental health awareness through multi-channel distribution and creative, narrative-driven designs.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 06:22:54 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:30:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yan",
"Samantha Bei Yi",
"",
"for the MINDLINE Study Group"
],
[
"Gunasekeran",
"Dinesh Visva",
"",
"for the MINDLINE Study Group"
],
[
"Tan",
"Caitlyn",
"",
"for the MINDLINE Study Group"
],
[
"Chan",
"Kai En",
"",
"for the MINDLINE Study Group"
],
[
"Tan",
"Caleb",
"",
"for the MINDLINE Study Group"
],
[
"Lim",
"Charmaine Shi Min",
"",
"for the MINDLINE Study Group"
],
[
"Chia",
"Audrey",
"",
"for the MINDLINE Study Group"
],
[
"Lei",
"Hsien-Hsien",
"",
"for the MINDLINE Study Group"
],
[
"Morris",
"Robert",
"",
"for the MINDLINE Study Group"
],
[
"Weng",
"Janice Huiqin",
"",
"for the MINDLINE Study Group"
]
] |
2508.20153
|
Alessandro Colombo
|
Linos Hecht, Andre Al Haddad, Bj\"orn Bastian, Thomas M. Baumann, Johan Bielecki, Christoph Bostedt, Subhendu De, Alberto De Fanis, Simon Dold, Thomas Fennel, Fanny Goy, Christina Graf, Robert Hartmann, Georg Jakobs, Maximilian Joschko, Gregor Knopp, Katharina Kolatzki, Sivarama Krishnan, Bj\"orn Kruse, Asbj{\o}rn {\O}. L{\ae}gdsmand, Bruno Langbehn, Suddhasattwa Mandal, Tommaso Mazza, Michael Meyer, Christian Peltz, Thomas Pfeifer, Safi Rafie-Zinedine, Antoine Sarracini, Mario Sauppe, Florian Schenk, Kirsten Schnorr, Bj\"orn Senfftleben, Keshav Sishodia, Frank Stienkemeier, Zhibin Sun, Rico Mayro P. Tanyag, Paul T\"ummler, Sergey Usenko, Carl Frederic Ussling, Vanessa Wood, Xinhua Xie, Maksym Yarema, Olesya Yarema, Nuri Yazdani, Hankai Zhang, Bernd von Issendorff, Yevheniy Ovcharenko, Marcel Mudrich, Daniela Rupp, Alessandro Colombo
|
Dichography: Two-frame Ultrafast Imaging from a Single Diffraction Pattern
|
19 pages, 7 figures. Supplemental Material in ancillary files. References and acknowledgements updated
| null | null | null |
physics.optics cond-mat.mes-hall physics.data-an
|
http://creativecommons.org/licenses/by/4.0/
|
X-ray free-electron lasers can nowadays deliver pairs of ultrabright, ultrashort light pulses with controllable delays at different photon energies. When combined with diffraction experiments, the interaction of these pulses with a sample at different times can produce two snapshots of an evolving system, enabling spatial and temporal resolution of ultrafast dynamics in nanomatter at and beyond the terahertz time scale. However, light detectors are orders of magnitude slower. The diffraction signals from the two pulses overlap in the recorded data and cannot be separated using currently available analysis methods. Here, we address this challenge by introducing Dichography, a diffraction imaging technique that restores the two unique views of the sample from superimposed scattering signals. We apply Dichography to experimental diffraction patterns of isolated xenon-doped superfluid helium nanodroplets, imaged using time-delayed, two-color X-ray pulses at intensities near the detection limit. The reconstructed images provide evidence of the survival of the xenon structures up to 750 fs after the interaction with the first shot. The capabilities of Dichography are further explored by applying the method to data from a second experiment, in which we retrieve the images of two distinct silver nanoparticles intercepted by the same light pulse. Dichography enables a new class of experiments for spatially and temporally resolving ultrafast phenomena, making a significant step toward the original promise of X-ray free-electron lasers to capture ultrafast movies of nanomatter.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 17:49:06 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 12:45:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hecht",
"Linos",
""
],
[
"Haddad",
"Andre Al",
""
],
[
"Bastian",
"Björn",
""
],
[
"Baumann",
"Thomas M.",
""
],
[
"Bielecki",
"Johan",
""
],
[
"Bostedt",
"Christoph",
""
],
[
"De",
"Subhendu",
""
],
[
"De Fanis",
"Alberto",
""
],
[
"Dold",
"Simon",
""
],
[
"Fennel",
"Thomas",
""
],
[
"Goy",
"Fanny",
""
],
[
"Graf",
"Christina",
""
],
[
"Hartmann",
"Robert",
""
],
[
"Jakobs",
"Georg",
""
],
[
"Joschko",
"Maximilian",
""
],
[
"Knopp",
"Gregor",
""
],
[
"Kolatzki",
"Katharina",
""
],
[
"Krishnan",
"Sivarama",
""
],
[
"Kruse",
"Björn",
""
],
[
"Lægdsmand",
"Asbjørn Ø.",
""
],
[
"Langbehn",
"Bruno",
""
],
[
"Mandal",
"Suddhasattwa",
""
],
[
"Mazza",
"Tommaso",
""
],
[
"Meyer",
"Michael",
""
],
[
"Peltz",
"Christian",
""
],
[
"Pfeifer",
"Thomas",
""
],
[
"Rafie-Zinedine",
"Safi",
""
],
[
"Sarracini",
"Antoine",
""
],
[
"Sauppe",
"Mario",
""
],
[
"Schenk",
"Florian",
""
],
[
"Schnorr",
"Kirsten",
""
],
[
"Senfftleben",
"Björn",
""
],
[
"Sishodia",
"Keshav",
""
],
[
"Stienkemeier",
"Frank",
""
],
[
"Sun",
"Zhibin",
""
],
[
"Tanyag",
"Rico Mayro P.",
""
],
[
"Tümmler",
"Paul",
""
],
[
"Usenko",
"Sergey",
""
],
[
"Ussling",
"Carl Frederic",
""
],
[
"Wood",
"Vanessa",
""
],
[
"Xie",
"Xinhua",
""
],
[
"Yarema",
"Maksym",
""
],
[
"Yarema",
"Olesya",
""
],
[
"Yazdani",
"Nuri",
""
],
[
"Zhang",
"Hankai",
""
],
[
"von Issendorff",
"Bernd",
""
],
[
"Ovcharenko",
"Yevheniy",
""
],
[
"Mudrich",
"Marcel",
""
],
[
"Rupp",
"Daniela",
""
],
[
"Colombo",
"Alessandro",
""
]
] |
2508.20164
|
Subir Sachdev
|
Pietro M. Bonetti and Maine Christos and Alexander Nikolaenko and Aavishkar A. Patel and Subir Sachdev
|
Critical quantum liquids and the cuprate high temperature superconductors
|
90 pages, 43 figures. Review article based on lectures by SS at Boulder, Trieste, Hong Kong, with links to lecture videos. Comments welcome
| null | null | null |
cond-mat.str-el hep-th
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
We present a theoretical framework for the cuprate superconductors, rooted in a fractionalized Fermi liquid (FL*) description of the intermediate-temperature pseudogap phase at low doping. The FL* theory predicted hole pockets each of fractional area $p/8$ at hole doping $p$, in contrast to the area $p/4$ in a spin density wave state or its thermal fluctuation. A recent magnetotransport observation of the Yamaji angle is in good agreement with area $p/8$.
We review a theory for the FL* phase of a single-band model using a pair of ancilla qubits on each site. Its mean field theory yields hole pockets of area $p/8$, and matches the gapped photoemission spectrum in the anti-nodal region of the Brillouin zone. Fluctuations are described by the SU(2) gauge theory of a background spin liquid with critical Dirac spinons. A Monte Carlo study of the thermal SU(2) gauge theory transforms the hole pockets into Fermi arcs. One route to confinement upon lowering temperature yields a $d$-wave superconductor via a Kosterlitz-Thouless transition of $h/(2e)$ vortices, with nodal Bogoliubov quasiparticles featuring anisotropic velocities and vortices surrounded by charge order halos. An alternative route produces a charge-ordered metallic state that exhibits quantum oscillations in agreement with experimental data.
Increasing doping from the FL* phase drives a transition to a conventional Fermi liquid at large doping, passing through an intermediate strange metal regime. We formulate a theory of this metal using a critical quantum `charge' liquid of mobile electrons in the presence of disorder, developed via an extension of the Sachdev-Ye-Kitaev model.
At low temperatures, and across optimal and over doping, we address the regimes of extended non-Fermi liquid behavior by Griffiths effects near quantum phase transitions in disordered metals.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 18:00:01 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 09:35:10 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Bonetti",
"Pietro M.",
""
],
[
"Christos",
"Maine",
""
],
[
"Nikolaenko",
"Alexander",
""
],
[
"Patel",
"Aavishkar A.",
""
],
[
"Sachdev",
"Subir",
""
]
] |
2508.20177
|
Kaila Ronayne
|
Kaila Ronayne, Casey Papovich, Allison Kirkpatrick, Bren E. Backhaus, Fergus Cullen, Lu Shen, Micaela B. Bagley, Guillermo Barro, Steven L. Finkelstein, Kurt Hamblin, Jeyhan S. Kartaltepe, Dale D. Kocevski, Anton M. Koekemoer, Erini Lambrides, Fabio Pacucci, Guang Yang
|
MEGA: Spectrophotometric SED Fitting of Little Red Dots Detected in JWST MIRI
|
34 pages, 14 figures, 6 tables, submitted to ApJ
| null | null | null |
astro-ph.GA
|
http://creativecommons.org/licenses/by/4.0/
|
We analyze eight spectroscopically confirmed Little Red Dots (LRDs) at redshifts $z = 5.1-8.7$ with JWST/NIRCam, NIRSpec, and MIRI data. The LRDs have red NIRCam colors, F150W-F444W $>$ 1, but flat NIRCam-MIRI colors, $-0.5 < \mathrm{F444W - F770W} < 0.5$, suggesting weak warm/hot dust components. The LRDs have $-1.0 < {F1000W - F1500W} < 1.1$, suggestive of non-uniform rest near-IR properties within the sample. We model the spectral energy distributions (SEDs) of the LRDs using the CIGALE and Prospector codes to assess how the differing templates impact the interpretation for LRDs for cases of: (1) models with star-forming stellar populations only; (2) active galactic nuclei (AGN) dominated models; and (3) composite AGN and star-forming models. Using the Bayesian information criterion, we find that six of the eight LRDs favor AGN models compared to star-forming models, though no model reproduces all of the observed properties. Two LRDs with pronounced Balmer-breaks and broad H$\alpha$ have SEDs that are reproduced with hot, dense-gas ($\log T/\mathrm{K}=5-5.7$, $\log n/\mathrm{cm^{-3}} = 9-11$) models with low dust attenuation ($A(V)\simeq 0.5$ mag). However, these models require an additional thermal component (800-1400 K) to account for the MIRI data, and fail to reproduce the rest-UV and narrow [OIII] emission. The total bolometric emission from the dense-gas models, and possibly CIGALE AGN models, appear consistent with literature constraints in the far-IR and radio, and require $\log L_{bol}/L_\odot<12$. These results suggest that our LRDs cannot be modeled entirely with standard templates, but instead require a novel treatment of gas conditions, AGN and star-formation.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 18:00:55 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 17:19:23 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Ronayne",
"Kaila",
""
],
[
"Papovich",
"Casey",
""
],
[
"Kirkpatrick",
"Allison",
""
],
[
"Backhaus",
"Bren E.",
""
],
[
"Cullen",
"Fergus",
""
],
[
"Shen",
"Lu",
""
],
[
"Bagley",
"Micaela B.",
""
],
[
"Barro",
"Guillermo",
""
],
[
"Finkelstein",
"Steven L.",
""
],
[
"Hamblin",
"Kurt",
""
],
[
"Kartaltepe",
"Jeyhan S.",
""
],
[
"Kocevski",
"Dale D.",
""
],
[
"Koekemoer",
"Anton M.",
""
],
[
"Lambrides",
"Erini",
""
],
[
"Pacucci",
"Fabio",
""
],
[
"Yang",
"Guang",
""
]
] |
2508.20250
|
Jessica Kinnevan
|
Jessica Kinnevan, Naifa Alqahtani, Toral Chauhan
|
Efficient and Privacy-Protecting Background Removal for 2D Video Streaming using iPhone 15 Pro Max LiDAR
| null | null | null | null |
eess.IV cs.CV cs.MM
|
http://creativecommons.org/licenses/by/4.0/
|
Light Detection and Ranging (LiDAR) technology in consumer-grade mobile devices can be used as a replacement for traditional background removal and compositing techniques. Unlike approaches such as chroma keying and trained AI models, LiDAR's depth information is independent of subject lighting, and performs equally well in low-light and well-lit environments. We integrate the LiDAR and color cameras on the iPhone 15 Pro Max with GPU-based image processing. We use Apple's SwiftUI and Swift frameworks for user interface and backend development, and Metal Shader Language (MSL) for realtime image enhancement at the standard iPhone streaming frame rate of 60 frames per second. The only meaningful limitations of the technology are the streaming bandwidth of the depth data, which currently reduces the depth map resolution to 320x240, and any pre-existing limitations of the LiDAR IR laser to reflect accurate depth from some materials. If the LiDAR resolution on a mobile device like the iPhone can be improved to match the color image resolution, LiDAR could feasibly become the preeminent method of background removal for video applications and photography.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 20:14:12 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 20:45:29 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Kinnevan",
"Jessica",
""
],
[
"Alqahtani",
"Naifa",
""
],
[
"Chauhan",
"Toral",
""
]
] |
2508.20299
|
Peter Pronko PhD
|
Peter P. Pronko and Paul A. Van Rompay
|
Isotope Enrichment and Element Separation by Self Generated Magnetic Centrifuge Fields in Ultrafast Laser Ablation
| null | null | null | null |
physics.plasm-ph physics.app-ph
|
http://creativecommons.org/publicdomain/zero/1.0/
|
A self-consistent model is developed to explain the anomalously large enrichment of nickel isotopes observed in ablation plumes from ultrafast laser irradiation of solid surfaces. The model is based on the spontaneous creation of a magnetic centrifuge in the ablation plume and the associated cyclotron rotation of plasma ions with rotation rates on the order of 10^9 radians per second. Mass separation occurs around the radial coordinate of cylindrical symmetry with longitudinal axis normal to the ablating surface. A Gaussian shaped radial magnetic field distribution is extracted for Ni isotopes with peak field of 53 Megagauss and average field of 20 Megagauss. In addition to cyclotron rotation of ions, a rigid rotor model is also presented that is associated with the hydrodynamic rotation of the entire plasma and is shown to be of little consequence for the isotope enrichment. Cyclotron rotations dominate the process. A profound resonance of enrichment is observed for specific cyclotron charge states and is shown to be associated with Ion Bernstein Wave (IBW) coupling that provides electrostatic acceleration to these resonant orbits.
|
[
{
"version": "v1",
"created": "Wed, 27 Aug 2025 22:10:22 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 17:42:18 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Pronko",
"Peter P.",
""
],
[
"Van Rompay",
"Paul A.",
""
]
] |
2508.20330
|
Zohair Shafi
|
Zohair Shafi and Serdar Kadioglu
|
FORGE: Foundational Optimization Representations from Graph Embeddings
| null | null | null | null |
cs.LG
|
http://creativecommons.org/licenses/by/4.0/
|
Combinatorial optimization problems are ubiquitous in science and engineering, yet learning-based approaches to accelerate their solution often require solving a large number of hard-to-solve optimization instances to collect training data, incurring significant computational overhead. Existing methods require training dedicated models for each problem distribution for each downstream task, severely limiting their scalability and generalization. In this work, we introduce Forge, a method of pre-training a vector-quantized graph autoencoder on a large and diverse collection of mixed-integer programming (MIP) instances in an unsupervised fashion without dependency on their solution. The vector quantization process creates discrete code assignments that act as a vocabulary to represent optimization instances. We evaluate our approach under both supervised and unsupervised settings. For the unsupervised setting, we demonstrate that Forge embeddings effectively differentiate and cluster unseen instances. For the supervised setting, we fine-tuneForge embeddings and show that a single model predicts both the variables for warm-starts and integrality gaps for cut-generation across multiple problem type distributions. Both predictions help improve performance of a state-of-the-art, commercial optimization solver. Finally, we release our code and pre-trained Forge weights to encourage further research and practical use of instance-level MIP embeddings at https://github.com/skadio/forge/.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 00:15:57 GMT"
},
{
"version": "v2",
"created": "Fri, 29 Aug 2025 16:55:40 GMT"
},
{
"version": "v3",
"created": "Mon, 01 Sep 2025 01:01:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Shafi",
"Zohair",
""
],
[
"Kadioglu",
"Serdar",
""
]
] |
2508.20386
|
Ao Chen
|
Ao Chen, Juntai Shen, Chun Wang, Yang Huang
|
Revisiting the Radial Metallicity Gradient-Age Relation in the Milky Way's Thin and Thick Disks
|
13 pages, 9 figures, submitted to ApJ
| null | null | null |
astro-ph.GA
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Galactic disks typically exhibit a negative radial metallicity gradient, indicating faster enrichment in the inner regions. Recent studies report that this gradient becomes flatter with increasing stellar age in the Milky Way's (MW) thin disk, while the thick disk exhibits a mildly positive gradient across all ages. In this work, we revisit the metallicity gradient-age relation (MGAR) in both the thin and thick disks of the MW. We use spectroscopic data from LAMOST DR8 and stellar ages calibrated with asteroseismology. Our results show a steadily flattening MGAR in the thin disk and confirm a positive gradient $\sim0.013\,\mathrm{dex\,kpc^{-1}}$ in the thick disk. The flattening in the thin disk may be caused by large-scale radial migration induced by transient spiral arms, or by a time-dependent steepening of the interstellar medium (ISM) metallicity gradient as suggested by recent FIRE2 simulations. The positive gradient in the thick disk may reflect early enrichment of the outer regions by strong feedback or starburst-driven outflows in a turbulent, gas-rich proto-disk. These findings suggest distinct chemodynamical evolution paths for the MW's thin and thick disks and provide valuable constraints for future models of Galactic chemical evolution.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 03:20:02 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 06:25:50 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Chen",
"Ao",
""
],
[
"Shen",
"Juntai",
""
],
[
"Wang",
"Chun",
""
],
[
"Huang",
"Yang",
""
]
] |
2508.20404
|
Chenyi Zhuang
|
Chengyue Yu, Siyuan Lu, Chenyi Zhuang, Dong Wang, Qintong Wu, Zongyue Li, Runsheng Gan, Chunfeng Wang, Siqi Hou, Gaochi Huang, Wenlong Yan, Lifeng Hong, Aohui Xue, Yanfeng Wang, Jinjie Gu, David Tsai, Tao Lin
|
AWorld: Orchestrating the Training Recipe for Agentic AI
| null | null | null | null |
cs.AI
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
The learning from practice paradigm is crucial for developing capable Agentic AI systems, yet it is severely hampered by inefficient experience generation, a bottleneck especially pronounced in complex benchmarks like GAIA. To address this, we introduce AWorld, an open-source system engineered for large-scale agent-environment interaction. By distributing tasks across a cluster, AWorld accelerates experience collection by 14.6x compared to standard single-node, sequential execution. This critical speedup makes extensive reinforcement learning practical and scalable. Leveraging this capability, we trained a Qwen3-32B-based agent that achieves pass@1 accuracy of 32.23% on the GAIA test set, which surpasses GPT-4o (27.91%) and rivals DeepSeek-V3 (31.89%). Our open-source system and the resulting agent provide a practical blueprint for a complete agentic AI training pipeline, from efficient interaction to demonstrable model improvement.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 04:04:30 GMT"
},
{
"version": "v2",
"created": "Mon, 01 Sep 2025 03:56:31 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Yu",
"Chengyue",
""
],
[
"Lu",
"Siyuan",
""
],
[
"Zhuang",
"Chenyi",
""
],
[
"Wang",
"Dong",
""
],
[
"Wu",
"Qintong",
""
],
[
"Li",
"Zongyue",
""
],
[
"Gan",
"Runsheng",
""
],
[
"Wang",
"Chunfeng",
""
],
[
"Hou",
"Siqi",
""
],
[
"Huang",
"Gaochi",
""
],
[
"Yan",
"Wenlong",
""
],
[
"Hong",
"Lifeng",
""
],
[
"Xue",
"Aohui",
""
],
[
"Wang",
"Yanfeng",
""
],
[
"Gu",
"Jinjie",
""
],
[
"Tsai",
"David",
""
],
[
"Lin",
"Tao",
""
]
] |
2508.20418
|
Siddharth Mahesh
|
Siddharth Mahesh and Sean T. McWilliams and Zachariah Etienne
|
Spinning Effective-to-Backwards One Body ($\texttt{SEBOB}$): combining Effective One-Body inspirals and Backwards One-Body merger-ringdowns for aligned spin black hole binaries
|
17 pages, 9 figures
| null | null | null |
gr-qc
|
http://creativecommons.org/licenses/by/4.0/
|
High-fidelity gravitational waveform models are essential for realizing the scientific potential of next-generation gravitational-wave observatories. While highly accurate, state-of-the-art models often rely on extensive phenomenological calibrations to numerical relativity (NR) simulations for the late-inspiral and merger phases, which can limit physical insight and extrapolation to regions where NR data is sparse. To address this, we introduce the Spinning Effective-to-Backwards One Body (SEBOB) formalism, a hybrid approach that combines the well-established Effective-One-Body (EOB) framework with the analytically-driven Backwards-One-Body (BOB) model, which describes the merger-ringdown from first principles as a perturbation of the final remnant black hole. We present two variants building on the state-of-the-art $\texttt{SEOBNRv5HM}$ model: $\texttt{seobnrv5_nrnqc_bob}$, which retains standard NR-calibrated non-quasi-circular (NQC) corrections and attaches a BOB-based merger-ringdown; and a more ambitious variant, $\texttt{seobnrv5_bob}$, which uses BOB to also inform the NQC corrections, thereby reducing reliance on NR fitting and enabling higher-order ($\mathcal{C}^2$) continuity by construction. Implemented in the open-source $\texttt{NRPy}$ framework for optimized C-code generation, the SEBOB model is transparent, extensible, and computationally efficient. By comparing our waveforms to a large catalog of NR simulations, we demonstrate that SEBOB yields accuracies comparable to the highly-calibrated $\texttt{SEOBNRv5HM}$ model, providing a viable pathway towards more physically motivated and robust waveform models
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 04:37:19 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 18:51:37 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mahesh",
"Siddharth",
""
],
[
"McWilliams",
"Sean T.",
""
],
[
"Etienne",
"Zachariah",
""
]
] |
2508.20425
|
Samuel Dayo
|
Samuel Dayo, Shuhan Liu, Peijing Li, Philip Levis, Subhasish Mitra, Thierry Tambe, David Tennenhouse, and H.-S. Philip Wong
|
The Future of Memory: Limits and Opportunities
|
3 Pages, 2 Figures, 1 Table, Accepted to SOSP 25 BigMem Workshop
| null | null | null |
cs.AR
|
http://creativecommons.org/licenses/by/4.0/
|
Memory latency, bandwidth, capacity, and energy increasingly limit performance. In this paper, we reconsider proposed system architectures that consist of huge (many-terabyte to petabyte scale) memories shared among large numbers of CPUs. We argue two practical engineering challenges, scaling and signaling, limit such designs. We propose the opposite approach. Rather than create large, shared, homogenous memories, systems explicitly break memory up into smaller slices more tightly coupled with compute elements. Leveraging advances in 2.5D/3D integration, this compute-memory node provisions private local memory, enabling accesses of node-exclusive data through micrometer-scale distances, and dramatically reduced access cost. In-package memory elements support shared state within a processor, providing far better bandwidth and energy-efficiency than DRAM, which is used as main memory for large working sets and cold data. Hardware making memory capacities and distances explicit allows software to efficiently compose this hierarchy, managing data placement and movement.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 04:46:57 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 20:43:11 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Dayo",
"Samuel",
""
],
[
"Liu",
"Shuhan",
""
],
[
"Li",
"Peijing",
""
],
[
"Levis",
"Philip",
""
],
[
"Mitra",
"Subhasish",
""
],
[
"Tambe",
"Thierry",
""
],
[
"Tennenhouse",
"David",
""
],
[
"Wong",
"H. -S. Philip",
""
]
] |
2508.20499
|
Zhen-Tao He
|
Zhen-Tao He, Jia Du, Jiageng Jiao, Caiying Shao, Junxi Shi, Yu Tian and Hongbao Zhang
|
Numerical computation of electromagnetically sourced nonlinear tails
|
20 pages, 10 figures; v2: minor improvements and eq. (57) is corrected
| null | null | null |
gr-qc astro-ph.HE
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Amazingly, recent studies indicate that nonlinear effects are of great significance for modelling black hole ringdown. Transient electromagnetic events in the astrophysical environment are typically high-energetic, potentially responsible for some nonlinearities in ringdown. Motivated by the desire to understand these nonlinearities, we solve the inhomogeneous Bardeen-Press-Teukolsky equation numerically, and find second-order gravitational tails induced by an electromagnetic source. Our results suggest that the second-order tails of curvature perturbations with multipole numbers $l\geq4$ decay as $t^{-2l-2}$ at fixed spatial position and $u^{-l-3}$ in retarded-time $u$ at null infinity, slower than their linear counterparts, which can play a role in multi-messenger observations.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 07:28:47 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 11:03:07 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"He",
"Zhen-Tao",
""
],
[
"Du",
"Jia",
""
],
[
"Jiao",
"Jiageng",
""
],
[
"Shao",
"Caiying",
""
],
[
"Shi",
"Junxi",
""
],
[
"Tian",
"Yu",
""
],
[
"Zhang",
"Hongbao",
""
]
] |
2508.20509
|
Yang Yang
|
Qingchun Hao, Yang Yang
|
Polynomial properties of unipotent radicals of parabolic subgroups in classical groups
|
17pages
| null | null | null |
math.GR math.CO math.RT
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Let $R_u^{X,d}$ denote the unipotent radical of a (proper) maximal standard parabolic subgroup of the classical group $\mathrm{Sp}_{2n}(q)$, $\mathrm{SO}_{2n}(q)$, or $\mathrm U_{2n}({q^2})$. This paper establishes explicit formulas for the number of irreducible characters of $R_u^{X,d}$ with degree $q^e$.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 07:49:08 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 08:30:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Hao",
"Qingchun",
""
],
[
"Yang",
"Yang",
""
]
] |
2508.20538
|
Georgy Lukyanov
|
Georgy Lukyanov and Darina Cheredina
|
False Cascades and the Cost of Truth
| null | null | null | null |
econ.TH
|
http://creativecommons.org/licenses/by/4.0/
|
We study sequential social learning when agents can sometimes pay to verify a claim and obtain hard, publicly checkable evidence. Each agent observes the public history, receives a private signal, may investigate at a cost (succeeding only when the claim is true), and can disclose or conceal any proof. Actions are binary or continuous, with a conformity pull toward the prevailing consensus. We characterize when false cascades persist and when societies self-correct. In the binary benchmark, we derive an investigation cutoff and show how its location relative to classic cascade bands governs breakability; a simple knife-edge condition guarantees that any wrong cascade at the boundary is overturned with positive probability. With continuous actions, coarse observation and conformity can recreate cascades, yet occasional disclosures collapse them. These forces yield a tractable "resilience frontier" with transparent comparative statics and policy levers.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 08:27:55 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 17:47:02 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Lukyanov",
"Georgy",
""
],
[
"Cheredina",
"Darina",
""
]
] |
2508.20547
|
Yunpeng Mei
|
Yunpeng Mei, Hongjie Cao, Yinqiu Xia, Wei Xiao, Zhaohan Feng, Gang Wang, Jie Chen
|
SPGrasp: Spatiotemporal Prompt-driven Grasp Synthesis in Dynamic Scenes
| null | null | null | null |
cs.RO cs.AI cs.CV
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
Real-time interactive grasp synthesis for dynamic objects remains challenging as existing methods fail to achieve low-latency inference while maintaining promptability. To bridge this gap, we propose SPGrasp (spatiotemporal prompt-driven dynamic grasp synthesis), a novel framework extending segment anything model v2 (SAMv2) for video stream grasp estimation. Our core innovation integrates user prompts with spatiotemporal context, enabling real-time interaction with end-to-end latency as low as 59 ms while ensuring temporal consistency for dynamic objects. In benchmark evaluations, SPGrasp achieves instance-level grasp accuracies of 90.6% on OCID and 93.8% on Jacquard. On the challenging GraspNet-1Billion dataset under continuous tracking, SPGrasp achieves 92.0% accuracy with 73.1 ms per-frame latency, representing a 58.5% reduction compared to the prior state-of-the-art promptable method RoG-SAM while maintaining competitive accuracy. Real-world experiments involving 13 moving objects demonstrate a 94.8% success rate in interactive grasping scenarios. These results confirm SPGrasp effectively resolves the latency-interactivity trade-off in dynamic grasp synthesis.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 08:38:50 GMT"
},
{
"version": "v2",
"created": "Sat, 30 Aug 2025 05:51:05 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Mei",
"Yunpeng",
""
],
[
"Cao",
"Hongjie",
""
],
[
"Xia",
"Yinqiu",
""
],
[
"Xiao",
"Wei",
""
],
[
"Feng",
"Zhaohan",
""
],
[
"Wang",
"Gang",
""
],
[
"Chen",
"Jie",
""
]
] |
2508.20634
|
Pradip Kumar Chattopadhyay Dr.
|
Koushik Ballav Goswami, Debadri Bhattacharjee, Anirban Saha, and Pradip Kumar Chattopadhyay
|
Exploring density dependent B as a suitable parameter in higher dimensional approach with a non-linear equation of state
|
23 pages, 21 figures
| null | null | null |
gr-qc
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
In this investigation, we present a singularity free interior solution of the Einstein field equation for a class of anisotropic compact objects in dimensions $D\geq4$. In accordance with the concept of Vaidya and Tikekar, the geometry of the physical $(D-1)$-space of a star corresponding to $t=constant$ hypersurface is assumed to be of a $(D-1)$ spheroid. For the fulfilment of causality condition, a limit of the spheroidal parameter ($\lambda$) is noted depending on the values of amount of anisotropy ($\alpha$) and space-time dimensions ($D$). We note that by switching off the extra parameters ($\alpha$ and $D$), previously obtained limit of $\lambda$ can be generated. To validate our findings, we compare the results obtained from our model with observational data of PSR J1614-2230 (mass=$1.908^{+0.016}_{-0.016}M_{\odot}$, radius=$11.93^{+0.50}_{-0.50}km$). It is noted that the best fit equation of state corresponds to polynomial equation of state of the order of five. We use this finding to develop a density dependent MIT bag model which seems to be useful for the correct description of compact object in our model. The mass radius relation shows that our model mimics a wide range of recently observed pulsars in four and higher dimensions. Furthermore, we also found that our model exhibits stability according to Generalised TOV equation, Herrera cracking condition, and the adiabatic index.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 10:33:50 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 15:17:53 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Goswami",
"Koushik Ballav",
""
],
[
"Bhattacharjee",
"Debadri",
""
],
[
"Saha",
"Anirban",
""
],
[
"Chattopadhyay",
"Pradip Kumar",
""
]
] |
2508.20760
|
Alma Liezenga
|
Jan Erik van Woerden, Gertjan Burghouts, Lotte Nijskens, Alma M. Liezenga, Sabina van Rooij, Frank Ruis, Hugo J. Kuijf
|
Occlusion Robustness of CLIP for Military Vehicle Classification
|
To be presented at SPIE: Sensors + Imaging, Artificial Intelligence for Security and Defence Applications II
| null | null | null |
cs.CV cs.AI
|
http://creativecommons.org/licenses/by/4.0/
|
Vision-language models (VLMs) like CLIP enable zero-shot classification by aligning images and text in a shared embedding space, offering advantages for defense applications with scarce labeled data. However, CLIP's robustness in challenging military environments, with partial occlusion and degraded signal-to-noise ratio (SNR), remains underexplored. We investigate CLIP variants' robustness to occlusion using a custom dataset of 18 military vehicle classes and evaluate using Normalized Area Under the Curve (NAUC) across occlusion percentages. Four key insights emerge: (1) Transformer-based CLIP models consistently outperform CNNs, (2) fine-grained, dispersed occlusions degrade performance more than larger contiguous occlusions, (3) despite improved accuracy, performance of linear-probed models sharply drops at around 35% occlusion, (4) by finetuning the model's backbone, this performance drop occurs at more than 60% occlusion. These results underscore the importance of occlusion-specific augmentations during training and the need for further exploration into patch-level sensitivity and architectural resilience for real-world deployment of CLIP.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:16:55 GMT"
},
{
"version": "v2",
"created": "Tue, 02 Sep 2025 15:53:42 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"van Woerden",
"Jan Erik",
""
],
[
"Burghouts",
"Gertjan",
""
],
[
"Nijskens",
"Lotte",
""
],
[
"Liezenga",
"Alma M.",
""
],
[
"van Rooij",
"Sabina",
""
],
[
"Ruis",
"Frank",
""
],
[
"Kuijf",
"Hugo J.",
""
]
] |
2508.20778
|
Xinhao Huang
|
Xinhao Huang, Zhibo Ren, Yipeng Yu, Ying Zhou, Zulong Chen, Zeyi Wen
|
SEAL: Structure and Element Aware Learning to Improve Long Structured Document Retrieval
|
Accepted at EMNLP 2025 Main Conference
| null | null | null |
cs.IR cs.LG
|
http://arxiv.org/licenses/nonexclusive-distrib/1.0/
|
In long structured document retrieval, existing methods typically fine-tune pre-trained language models (PLMs) using contrastive learning on datasets lacking explicit structural information. This practice suffers from two critical issues: 1) current methods fail to leverage structural features and element-level semantics effectively, and 2) the lack of datasets containing structural metadata. To bridge these gaps, we propose \our, a novel contrastive learning framework. It leverages structure-aware learning to preserve semantic hierarchies and masked element alignment for fine-grained semantic discrimination. Furthermore, we release \dataset, a long structured document retrieval dataset with rich structural annotations. Extensive experiments on both released and industrial datasets across various modern PLMs, along with online A/B testing, demonstrate consistent performance improvements, boosting NDCG@10 from 73.96\% to 77.84\% on BGE-M3. The resources are available at https://github.com/xinhaoH/SEAL.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 13:34:42 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 09:48:40 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Huang",
"Xinhao",
""
],
[
"Ren",
"Zhibo",
""
],
[
"Yu",
"Yipeng",
""
],
[
"Zhou",
"Ying",
""
],
[
"Chen",
"Zulong",
""
],
[
"Wen",
"Zeyi",
""
]
] |
2508.20817
|
Xingchen Zhang
|
He Li, Xinyu Liu, Weihang Kong, Xingchen Zhang
|
FusionCounting: Robust visible-infrared image fusion guided by crowd counting via multi-task learning
|
11 pages, 9 figures
| null | null | null |
cs.CV
|
http://creativecommons.org/licenses/by-nc-nd/4.0/
|
Visible and infrared image fusion (VIF) is an important multimedia task in computer vision. Most VIF methods focus primarily on optimizing fused image quality. Recent studies have begun incorporating downstream tasks, such as semantic segmentation and object detection, to provide semantic guidance for VIF. However, semantic segmentation requires extensive annotations, while object detection, despite reducing annotation efforts compared with segmentation, faces challenges in highly crowded scenes due to overlapping bounding boxes and occlusion. Moreover, although RGB-T crowd counting has gained increasing attention in recent years, no studies have integrated VIF and crowd counting into a unified framework. To address these challenges, we propose FusionCounting, a novel multi-task learning framework that integrates crowd counting into the VIF process. Crowd counting provides a direct quantitative measure of population density with minimal annotation, making it particularly suitable for dense scenes. Our framework leverages both input images and population density information in a mutually beneficial multi-task design. To accelerate convergence and balance tasks contributions, we introduce a dynamic loss function weighting strategy. Furthermore, we incorporate adversarial training to enhance the robustness of both VIF and crowd counting, improving the model's stability and resilience to adversarial attacks. Experimental results on public datasets demonstrate that FusionCounting not only enhances image fusion quality but also achieves superior crowd counting performance.
|
[
{
"version": "v1",
"created": "Thu, 28 Aug 2025 14:15:18 GMT"
},
{
"version": "v2",
"created": "Sun, 31 Aug 2025 17:52:52 GMT"
}
] | 2025-09-03T00:00:00 |
[
[
"Li",
"He",
""
],
[
"Liu",
"Xinyu",
""
],
[
"Kong",
"Weihang",
""
],
[
"Zhang",
"Xingchen",
""
]
] |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.