# Research Papers — 2025-10 · bundle page 216 of 217

> Full-text markdown of papers #10751–10800 of 10843 published in 2025-10, newest first. AI-readable mirrors of the full 10843-paper 2025-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2025-10/page/216
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/215.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/217.md)

### EMR-AGENT: Automating Cohort and Feature Extraction from EMR Databases

Authors: Kwanhyung Lee, Sungsoo Hong, Joonhyung Park, Jeonghyeop Lim, Juhwan Choi, Donghwee Yoon, Eunho Yang
Published: 2025-10-01
Categories: cs.DB, cs.AI
arXiv: https://arxiv.org/abs/2510.00549v2

Machine learning models for clinical prediction rely on structured data extracted from Electronic Medical Records (EMRs), yet this process remains dominated by hardcoded, database-specific pipelines for cohort definition, feature selection, and code mapping. These manual efforts limit scalability, reproducibility, and cross-institutional generalization. To address this, we introduce EMR-AGENT (Automated Generalized Extraction and Navigation Tool), an agent-based framework that replaces manual rule writing with dynamic, language model-driven interaction to extract and standardize structured clinical data. Our framework automates cohort selection, feature extraction, and code mapping through interactive querying of databases. Our modular agents iteratively observe query results and reason over schema and documentation, using SQL not just for data retrieval but also as a tool for database observation and decision making. This eliminates the need for hand-crafted, schema-specific logic. To enable rigorous evaluation, we develop a benchmarking codebase for three EMR databases (MIMIC-III, eICU, SICdb), including both seen and unseen schema settings. Our results demonstrate strong performance and generalization across these databases, highlighting the feasibility of automating a process previously thought to require expert-driven design. The code will be released publicly at https://github.com/AITRICS/EMR-AGENT/tree/main. For a demonstration, please visit our anonymous demo page: https://anonymoususer-max600.github.io/EMR_AGENT/

---

### Forestpest-YOLO: A High-Performance Detection Framework for Small Forestry Pests

Authors: Aoduo Li, Peikai Lin, Jiancheng Li, Zhen Zhang, Shiting Wu, Zexiao Liang, Zhifa Jiang
Published: 2025-10-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.00547v1

Detecting agricultural pests in complex forestry environments using remote sensing imagery is fundamental for ecological preservation, yet it is severely hampered by practical challenges. Targets are often minuscule, heavily occluded, and visually similar to the cluttered background, causing conventional object detection models to falter due to the loss of fine-grained features and an inability to handle extreme data imbalance. To overcome these obstacles, this paper introduces Forestpest-YOLO, a detection framework meticulously optimized for the nuances of forestry remote sensing. Building upon the YOLOv8 architecture, our framework introduces a synergistic trio of innovations. We first integrate a lossless downsampling module, SPD-Conv, to ensure that critical high-resolution details of small targets are preserved throughout the network. This is complemented by a novel cross-stage feature fusion block, CSPOK, which dynamically enhances multi-scale feature representation while suppressing background noise. Finally, we employ VarifocalLoss to refine the training objective, compelling the model to focus on high-quality and hard-to-classify samples. Extensive experiments on our challenging, self-constructed ForestPest dataset demonstrate that Forestpest-YOLO achieves state-of-the-art performance, showing marked improvements in detecting small, occluded pests and significantly outperforming established baseline models.

---

### ThinkBrake: Efficient Reasoning via Log-Probability Margin Guided Decoding

Authors: Sangjun Song, Minjae Oh, Seungkyu Lee, Sungmin Jo, Yohan Jo
Published: 2025-10-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.00546v5

Large Reasoning Models (LRMs) allocate substantial inference-time compute to Chain-of-Thought (CoT) reasoning, improving performance on mathematics, scientific QA, and tool usage. However, this introduces overthinking: LRMs often reach a correct intermediate solution, continue reasoning, and overwrite it with an incorrect answer. We first demonstrate that oracle stopping--where we inject </think> at every sentence boundary and select the best stopping point in hindsight--improves average accuracy by 8% while reducing thinking tokens by 72%, exposing substantial overthinking. Motivated by this finding, we propose ThinkBrake, which monitors the log-probability margin between the top continuation token and </think> at sentence boundaries, stopping reasoning when this margin narrows. ThinkBrake requires no training and achieves favorable accuracy-efficiency trade-offs across math, scientific QA, and tool usage benchmarks, reducing thinking token usage by up to 30%. Furthermore, we provide theoretical analysis showing that ThinkBrake is equivalent to test-time realignment with a reward bonus for the </think> token.

---

### Bayesian Neural Networks for Functional ANOVA model

Authors: Seokhun Park, Choeun Kim, Jihu Lee, Yunseop Shin, Insung Kong, Yongdai Kim
Published: 2025-10-01
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.00545v2

With the increasing demand for interpretability in machine learning, functional ANOVA decomposition has gained renewed attention as a principled tool for breaking down high-dimensional function into low-dimensional components that reveal the contributions of different variable groups. Recently, Tensor Product Neural Network (TPNN) has been developed and applied as basis functions in the functional ANOVA model, referred to as ANOVA-TPNN. A disadvantage of ANOVA-TPNN, however, is that the components to be estimated must be specified in advance, which makes it difficult to incorporate higher-order TPNNs into the functional ANOVA model due to computational and memory constraints. In this work, we propose Bayesian-TPNN, a Bayesian inference procedure for the functional ANOVA model with TPNN basis functions, enabling the detection of higher-order components with reduced computational cost compared to ANOVA-TPNN. We develop an efficient MCMC algorithm and demonstrate that Bayesian-TPNN performs well by analyzing multiple benchmark datasets. Theoretically, we prove that the posterior of Bayesian-TPNN is consistent.

---

### Interpretable Machine Learning for Life Expectancy Prediction: A Comparative Study of Linear Regression, Decision Tree, and Random Forest

Authors: Roman Dolgopolyi, Ioanna Amaslidou, Agrippina Margaritou
Published: 2025-10-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.00542v1

Life expectancy is a fundamental indicator of population health and socio-economic well-being, yet accurately forecasting it remains challenging due to the interplay of demographic, environmental, and healthcare factors. This study evaluates three machine learning models -- Linear Regression (LR), Regression Decision Tree (RDT), and Random Forest (RF), using a real-world dataset drawn from World Health Organization (WHO) and United Nations (UN) sources. After extensive preprocessing to address missing values and inconsistencies, each model's performance was assessed with $R^2$, Mean Absolute Error (MAE), and Root Mean Squared Error (RMSE). Results show that RF achieves the highest predictive accuracy ($R^2 = 0.9423$), significantly outperforming LR and RDT. Interpretability was prioritized through p-values for LR and feature importance metrics for the tree-based models, revealing immunization rates (diphtheria, measles) and demographic attributes (HIV/AIDS, adult mortality) as critical drivers of life-expectancy predictions. These insights underscore the synergy between ensemble methods and transparency in addressing public-health challenges. Future research should explore advanced imputation strategies, alternative algorithms (e.g., neural networks), and updated data to further refine predictive accuracy and support evidence-based policymaking in global health contexts.

---

### Spectral Scaling Laws in Language Models: How Effectively Do Feed-Forward Networks Use Their Latent Space?

Authors: Nandan Kumar Jha, Brandon Reagen
Published: 2025-10-01
Categories: cs.LG, cs.IT
arXiv: https://arxiv.org/abs/2510.00537v1

As large language models (LLMs) scale, the question is not only how large they become, but how much of their capacity is effectively utilized. Existing scaling laws relate model size to loss, yet overlook how components exploit their latent space. We study feed-forward networks (FFNs) and recast width selection as a spectral utilization problem. Using a lightweight diagnostic suite -- Hard Rank (participation ratio), Soft Rank (Shannon rank), Spectral Concentration, and the composite Spectral Utilization Index (SUI) -- we quantify how many latent directions are meaningfully activated across LLaMA, GPT-2, and nGPT families. Our key finding is an asymmetric spectral scaling law: soft rank follows an almost perfect power law with FFN width, while hard rank grows only sublinearly and with high variance. This asymmetry suggests that widening FFNs mostly adds low-energy tail directions, while dominant-mode subspaces saturate early. Moreover, at larger widths, variance further collapses into a narrow subspace, leaving much of the latent space under-utilized. These results recast FFN width selection as a principled trade-off between tail capacity and dominant-mode capacity, offering concrete guidance for inference-efficient LLM design.

---

### GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness

Authors: Kung-Hsiang Huang, Haoyi Qiu, Yutong Dai, Caiming Xiong, Chien-Sheng Wu
Published: 2025-10-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.00536v2

Graphical user interface (GUI) agents built on vision-language models have emerged as a promising approach to automate human-computer workflows. However, they also face the inefficiency challenge as they process long sequences of high-resolution screenshots and solving long-horizon tasks, making inference slow, costly and memory-bound. While key-value (KV) caching can mitigate this, storing the full cache is prohibitive for image-heavy contexts. Existing cache-compression methods are sub-optimal as they do not account for the spatial and temporal redundancy of GUIs. In this work, we first analyze attention patterns in GUI agent workloads and find that, unlike in natural images, attention sparsity is uniformly high across all transformer layers. This insight motivates a simple uniform budget allocation strategy, which we show empirically outperforms more complex layer-varying schemes. Building on this, we introduce GUI-KV, a plug-and-play KV cache compression method for GUI agents that requires no retraining. GUI-KV combines two novel techniques: (i) spatial saliency guidance, which augments attention scores with the L2 norm of hidden states to better preserve semantically important visual tokens, and (ii) temporal redundancy scoring, which projects previous frames' keys onto the current frame's key subspace to preferentially prune redundant history. Across standard GUI agent benchmarks and models, GUI-KV outperforms competitive KV compression baselines, closely matching full-cache accuracy at modest budgets. Notably, in a 5-screenshot setting on the AgentNetBench benchmark, GUI-KV reduces decoding FLOPs by 38.9% while increasing step accuracy by 4.1% over the full-cache baseline. These results demonstrate that exploiting GUI-specific redundancies enables efficient and reliable agent performance.

---

### Cascaded Diffusion Framework for Probabilistic Coarse-to-Fine Hand Pose Estimation

Authors: Taeyun Woo, Jinah Park, Tae-Kyun Kim
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00527v2

Deterministic models for 3D hand pose reconstruction, whether single-staged or cascaded, struggle with pose ambiguities caused by self-occlusions and complex hand articulations. Existing cascaded approaches progressively refine pose predictions in a coarse-to-fine manner, but their deterministic nature prevents them from modeling pose uncertainty. Conversely, recent probabilistic methods capture pose distributions but are confined to single-stage estimation, often yielding inaccurate 3D reconstructions without refinement. To address these limitations, we propose a coarse-to-fine cascaded diffusion framework that combines probabilistic modeling with cascaded refinement. The first stage is a joint diffusion model that samples diverse 3D joint hypotheses, and the second stage is a Mesh Latent Diffusion Model (Mesh LDM) that reconstructs a 3D hand mesh conditioned on a joint sample. Our key idea is to use these diverse hypotheses as a training signal rather than as final outputs, so that the Mesh LDM learns distribution-aware joint-mesh relationships and becomes robust to the variation in coarse predictions. Extensive ablations validate the necessity of the cascaded design and the choice of latent-space diffusion. Experiments on FreiHAND, HO3Dv2, and DexYCB show that our method achieves state-of-the-art performance and remains stable under occlusion and pose ambiguity.

---

### Beyond Log Likelihood: Probability-Based Objectives for Supervised Fine-Tuning across the Model Capability Continuum

Authors: Gaotang Li, Ruizhong Qiu, Xiusi Chen, Heng Ji, Hanghang Tong
Published: 2025-10-01
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.00526v3

Supervised fine-tuning (SFT) is the standard approach for post-training large language models (LLMs), yet it often shows limited generalization. We trace this limitation to its default training objective: negative log likelihood (NLL). While NLL is classically optimal when training from scratch, post-training operates in a different paradigm and could violate its optimality assumptions, where models already encode task-relevant priors and supervision can be long and noisy. In this work, we systematically study various probability-based objectives and characterize when and why different objectives succeed or fail under varying conditions. Through comprehensive experiments and extensive ablation studies across 8 model backbones, 27 benchmarks, and 7 domains, we uncover a critical dimension that governs objective behavior: the model-capability continuum. Near the model-strong end, prior-leaning objectives that downweight low-probability tokens (e.g., $-p$, $-p^{10}$, thresholded variants) consistently outperform NLL; toward the model-weak end, NLL dominates; in between, no single objective prevails. Our theoretical analysis further elucidates how objectives trade places across the continuum, providing a principled foundation for adapting objectives to model capability. The code is available at https://github.com/GaotangLi/Beyond-Log-Likelihood.

---

### VIRTUE: Visual-Interactive Text-Image Universal Embedder

Authors: Wei-Yao Wang, Kazuya Tateishi, Qiyu Wu, Shusuke Takahashi, Yuki Mitsufuji
Published: 2025-10-01
Categories: cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2510.00523v2

Multimodal representation learning models have demonstrated successful operation across complex tasks, and the integration of vision-language models (VLMs) has further enabled embedding models with instruction-following capabilities. However, existing embedding models lack visual-interactive capabilities to specify regions of interest from users (e.g., point, bounding box, mask), which have been explored in generative models to broaden their human-interactive applicability. Equipping embedding models with visual interactions not only would unlock new applications with localized grounding of user intent, which remains unexplored, but also enable the models to learn entity-level information within images to complement their global representations for conventional embedding tasks. In this paper, we propose a novel Visual-InteRactive Text-Image Universal Embedder (VIRTUE) that extends the capabilities of the segmentation model and the vision-language model to the realm of representation learning. In VIRTUE, the segmentation model can process visual prompts that pinpoint specific regions within an image, thereby enabling the embedder to handle complex and ambiguous scenarios more precisely. To evaluate the visual-interaction ability of VIRTUE, we introduce a large-scale Segmentation-and-Scene Caption Retrieval (SCaR) benchmark comprising 1M samples that aims to retrieve the text caption by jointly considering the entity with a specific object and image scene. VIRTUE consistently achieves a state-of-the-art performance with significant improvements across 36 universal MMEB (3.1%-8.5%) and five visual-interactive SCaR (15.2%-20.3%) tasks.

---

### CardioBench: Do Echocardiography Foundation Models Generalize Beyond the Lab?

Authors: Darya Taratynova, Ahmed Aly, Numan Saeed, Mohammad Yaqub
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00520v2

Foundation models are reshaping medical imaging, yet their application in echocardiography remains limited, hindered by a heavy reliance on private datasets that prevent reproducible comparison. Echocardiography poses unique challenges, including noisy acquisitions, high frame redundancy, and limited diverse public datasets. To address this, we introduce CardioBench, a comprehensive benchmark for echocardiography foundation models. Specifically, CardioBench unifies eight publicly available datasets into a standardized suite spanning four regression and five classification tasks, covering functional, structural, diagnostic, and view recognition endpoints. Leveraging this framework, we evaluate several leading foundation models, including cardiac-specific, biomedical, and general-purpose encoders, under consistent zero-shot, probing, and alignment protocols. Our analysis reveals that while general-purpose encoders transfer well and often close the gap with probing, they struggle significantly with fine-grained distinctions like view classification and subtle pathology recognition. Results indicate that models capturing temporal cardiac dynamics perform best on functional tasks, while retrieval-based approaches generalize more consistently across datasets. By releasing preprocessing, splits, and public evaluation pipelines, CardioBench establishes a reproducible reference point to guide the architectural design of future echocardiography and possibly other medical imaging foundation models.

---

### Architectural Transformations and Emerging Verification Demands in AI-Enabled Cyber-Physical Systems

Authors: Hadiza Umar Yusuf, Khouloud Gaaloul
Published: 2025-10-01
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2510.00519v1

In the world of Cyber-Physical Systems (CPS), a captivating real-time fusion occurs where digital technology meets the physical world. This synergy has been significantly transformed by the integration of artificial intelligence (AI), a move that dramatically enhances system adaptability and introduces a layer of complexity that impacts CPS control optimization and reliability. Despite advancements in AI integration, a significant gap remains in understanding how this shift affects CPS architecture, operational complexity, and verification practices. The extended abstract addresses this gap by investigating architectural distinctions between AI-driven and traditional control models designed in Simulink and their respective implications for system verification.

---

### Understanding Sensitivity of Differential Attention through the Lens of Adversarial Robustness

Authors: Tsubasa Takahashi, Shojiro Yamabe, Futa Waseda, Kento Sasaki
Published: 2025-10-01
Categories: cs.LG, cs.CR
arXiv: https://arxiv.org/abs/2510.00517v2

Differential Attention (DA) has been proposed as a refinement to standard attention, suppressing redundant or noisy context through a subtractive structure and thereby reducing contextual hallucination. While this design sharpens task-relevant focus, we show that it also introduces a structural fragility under adversarial perturbations. Our theoretical analysis identifies negative gradient alignment-a configuration encouraged by DA's subtraction-as the key driver of sensitivity amplification, leading to increased gradient norms and elevated local Lipschitz constants. We empirically validate this Fragile Principle through systematic experiments on ViT/DiffViT and evaluations of pretrained CLIP/DiffCLIP, spanning five datasets in total. These results demonstrate higher attack success rates, frequent gradient opposition, and stronger local sensitivity compared to standard attention. Furthermore, depth-dependent experiments reveal a robustness crossover: stacking DA layers attenuates small perturbations via depth-dependent noise cancellation, though this protection fades under larger attack budgets. Overall, our findings uncover a fundamental trade-off: DA improves discriminative focus on clean inputs but increases adversarial vulnerability, underscoring the need to jointly design for selectivity and robustness in future attention mechanisms.

---

### Efficient Multi-modal Large Language Models via Progressive Consistency Distillation

Authors: Zichen Wen, Shaobo Wang, Yufa Zhou, Junyuan Zhang, Qintong Zhang, Yifeng Gao, Zhaorun Chen, Bin Wang, Weijia Li, Conghui He, Linfeng Zhang
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00515v1

Visual tokens consume substantial computational resources in multi-modal large models (MLLMs), significantly compromising their efficiency. Recent works have attempted to improve efficiency by compressing visual tokens during training, either through modifications to model components or by introducing additional parameters. However, they often overlook the increased learning difficulty caused by such compression, as the model's parameter space struggles to quickly adapt to the substantial perturbations in the feature space induced by token compression. In this work, we propose to develop Efficient MLLMs via Progressive Consistency Distillation (EPIC), a progressive learning framework. Specifically, by decomposing the feature space perturbations introduced by token compression along the token-wise and layer-wise dimensions, we introduce token consistency distillation and layer consistency distillation, respectively, aiming to reduce the training difficulty by leveraging guidance from a teacher model and following a progressive learning trajectory. Extensive experiments demonstrate the superior effectiveness, robustness, and generalization capabilities of our proposed framework.

---

### EuroSpeech: A Multilingual Speech Corpus

Authors: Samuel Pfisterer, Florian Grötschla, Luca A. Lanzendörfer, Florian Yan, Roger Wattenhofer
Published: 2025-10-01
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.00514v2

Recent progress in speech processing has highlighted that high-quality performance across languages requires substantial training data for each individual language. While existing multilingual datasets cover many languages, they often contain insufficient data for most languages. Thus, trained models perform poorly on the majority of the supported languages. Our work addresses this challenge by introducing a scalable pipeline for constructing speech datasets from parliamentary recordings. The proposed pipeline includes robust components for media retrieval and a two-stage alignment algorithm designed to handle non-verbatim transcripts and long-form audio. Applying this pipeline to recordings from 22 European parliaments, we extract over 61k hours of aligned speech segments, achieving substantial per-language coverage with 19 languages exceeding 1k hours and 22 languages exceeding 500 hours of high-quality speech data. We obtain an average 41.8\% reduction in word error rates over baselines when finetuning an existing ASR model on our dataset, demonstrating the usefulness of our approach.

---

### Adaptive Data-Knowledge Alignment in Genetic Perturbation Prediction

Authors: Yuanfang Xiang, Lun Ai
Published: 2025-10-01
Categories: q-bio.MN, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.00512v2

The transcriptional response to genetic perturbation reveals fundamental insights into complex cellular systems. While current approaches have made progress in predicting genetic perturbation responses, they provide limited biological understanding and cannot systematically refine existing knowledge. Overcoming these limitations requires an end-to-end integration of data-driven learning and existing knowledge. However, this integration is challenging due to inconsistencies between data and knowledge bases, such as noise, misannotation, and incompleteness. To address this challenge, we propose ALIGNED (Adaptive aLignment for Inconsistent Genetic kNowledgE and Data), a neuro-symbolic framework based on the Abductive Learning (ABL) paradigm. This end-to-end framework aligns neural and symbolic components and performs systematic knowledge refinement. We introduce a balanced consistency metric to evaluate the predictions' consistency against both data and knowledge. Our results show that ALIGNED outperforms state-of-the-art methods by achieving the highest balanced consistency, while also re-discovering biologically meaningful knowledge. Our work advances beyond existing methods to enable both the transparency and the evolution of mechanistic biological understanding.

---

### JoyAgent-JDGenie: Technical Report on the GAIA

Authors: Jiarun Liu, Shiyue Xu, Shangkun Liu, Yang Li, Wen Liu, Min Liu, Xiaoqing Zhou, Hanmin Wang, Shilin Jia, zhen Wang, Shaohua Tian, Hanhao Li et al.
Published: 2025-10-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.00510v1

Large Language Models are increasingly deployed as autonomous agents for complex real-world tasks, yet existing systems often focus on isolated improvements without a unifying design for robustness and adaptability. We propose a generalist agent architecture that integrates three core components: a collective multi-agent framework combining planning and execution agents with critic model voting, a hierarchical memory system spanning working, semantic, and procedural layers, and a refined tool suite for search, code execution, and multimodal parsing. Evaluated on a comprehensive benchmark, our framework consistently outperforms open-source baselines and approaches the performance of proprietary systems. These results demonstrate the importance of system-level integration and highlight a path toward scalable, resilient, and adaptive AI assistants capable of operating across diverse domains and tasks.

---

### Copy-Paste to Mitigate Large Language Model Hallucinations

Authors: Yongchao Long, Xian Wu, Yingying Zhang, Xianbin Wen, Yuxi Zhou, Shenda Hong
Published: 2025-10-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.00508v2

While Retrieval-Augmented Generation (RAG) enables large language models (LLMs) to generate contextually grounded responses, contextual faithfulness remains challenging as LLMs may not consistently trust provided context, leading to hallucinations that undermine reliability. We observe an inverse correlation between response copying degree and context-unfaithful hallucinations on RAGTruth, suggesting that higher copying degrees reduce hallucinations by fostering genuine contextual belief. We propose CopyPasteLLM, obtained through two-stage high-copying response preference training. We design three prompting methods to enhance copying degree, demonstrating that high-copying responses achieve superior contextual faithfulness and hallucination control. These approaches enable a fully automated pipeline that transforms generated responses into high-copying preference data for training CopyPasteLLM. On FaithEval, ConFiQA and PubMedQA, CopyPasteLLM achieves best performance in both counterfactual and original contexts, remarkably with 12.2% to 24.5% accuracy improvements on FaithEval over the best baseline, while requiring only 365 training samples -- 1/50th of baseline data. To elucidate CopyPasteLLM's effectiveness, we propose the Context-Parameter Copying Capturing algorithm. Interestingly, this reveals that CopyPasteLLM recalibrates reliance on internal parametric knowledge rather than external knowledge during generation. All codes are available at https://github.com/longyongchao/CopyPasteLLM

---

### Graph2Eval: Automatic Multimodal Task Generation for Agents via Knowledge Graphs

Authors: Yurun Chen, Xavier Hu, Yuhan Liu, Ziqi Wang, Zeyi Liao, Lin Chen, Feng Wei, Yuxi Qian, Bo Zheng, Keting Yin, Shengyu Zhang
Published: 2025-10-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.00507v3

As multimodal LLM-driven agents advance in autonomy and generalization, traditional static datasets face inherent scalability limitations and are insufficient for fully assessing their capabilities in increasingly complex and diverse tasks. Existing studies have attempted to generate agent tasks using LLMs, but due to the inherent hallucinations of LLMs and the lack of internal data relationship modeling, these tasks often exhibit semantic inconsistencies and solvability issues. To address these challenges, we introduce Graph2Eval, a knowledge-graph-driven framework for automated, scalable, and semantically grounded agent task generation. At its core, Graph2Eval leverages a knowledge graph built from heterogeneous external data sources as a structured task space, generating multimodal agent tasks through subgraph sampling and task construction guided by task templates and meta-path strategies. To further ensure task reliability, a multi-stage filtering pipeline based on node reachability analysis, LLM scoring, and similarity analysis ensures the diversity and solvability of the generated tasks. By unifying both RAG Agent and Web Agent scenarios, Graph2Eval enables efficient generation of multimodal document understanding tasks and multi-step web interaction tasks. We instantiate the framework with Graph2Eval-Bench, a curated dataset of 1,319 tasks spanning document understanding and web interaction scenarios. Extensive experiments show that, on average, Graph2Eval improves task semantic consistency by 20% and solvability by 17% over baselines, while Graph2Eval-Bench effectively distinguishes agent performance, offering a new perspective on agent evaluation.

---

### Affordance-Guided Diffusion Prior for 3D Hand Reconstruction

Authors: Naru Suzuki, Takehiko Ohkawa, Tatsuro Banno, Jihyun Lee, Ryosuke Furuta, Yoichi Sato
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00506v3

How can we reconstruct 3D hand poses when large portions of the hand are heavily occluded by itself or by objects? Humans often resolve such ambiguities by leveraging contextual knowledge -- such as affordances, where an object's shape and function suggest how the object is typically grasped. Inspired by this observation, we propose a generative prior for hand pose refinement guided by affordance-aware textual descriptions of hand-object interactions (HOI). Our method employs a diffusion-based generative model that learns the distribution of plausible hand poses conditioned on affordance descriptions, which are inferred from a large vision-language model (VLM). This enables the refinement of occluded regions into more accurate and functionally coherent hand poses. Extensive experiments on HOGraspNet, a 3D hand-affordance dataset with severe occlusions, demonstrate that our affordance-guided refinement significantly improves hand pose estimation over both recent regression methods and diffusion-based refinement lacking contextual reasoning.

---

### A Fast and Precise Method for Searching Rectangular Tumor Regions in Brain MR Images

Authors: Hidenori Takeshima, Shuki Maruyama
Published: 2025-10-01
Categories: eess.IV, cs.CV
arXiv: https://arxiv.org/abs/2510.00505v1

Purpose: To develop a fast and precise method for searching rectangular regions in brain tumor images. Methods: The authors propose a new method for searching rectangular tumor regions in brain MR images. The proposed method consisted of a segmentation network and a fast search method with a user-controllable search metric. As the segmentation network, the U-Net whose encoder was replaced by the EfficientNet was used. In the fast search method, summed-area tables were used for accelerating sums of voxels in rectangular regions. Use of the summed-area tables enabled exhaustive search of the 3D offset (3D full search). The search metric was designed for giving priority to cubes over oblongs, and assigning better values for higher tumor fractions even if they exceeded target tumor fractions. The proposed computation and metric were compared with those used in a conventional method using the Brain Tumor Image Segmentation dataset. Results: When the 3D full search was used, the proposed computation (8 seconds) was 100-500 times faster than the conventional computation (11-40 minutes). When the user-controllable parts of the search metrics were changed variously, the tumor fractions of the proposed metric were higher than those of the conventional metric. In addition, the conventional metric preferred oblongs whereas the proposed metric preferred cubes. Conclusion: The proposed method is promising for implementing fast and precise search of rectangular tumor regions, which is useful for brain tumor diagnosis using MRI systems. The proposed computation reduced processing times of the 3D full search, and the proposed metric improved the quality of the assigned rectangular tumor regions.

---

### A universal compression theory for lottery ticket hypothesis and neural scaling laws

Authors: Hong-Yi Wang, Di Luo, Tomaso Poggio, Isaac L. Chuang, Liu Ziyin
Published: 2025-10-01
Categories: stat.ML, cond-mat.dis-nn, cs.IT, cs.LG
arXiv: https://arxiv.org/abs/2510.00504v2

When training large-scale models, the performance typically scales with the number of parameters and the dataset size according to a slow power law. A fundamental theoretical and practical question is whether comparable performance can be achieved with significantly smaller models and substantially less data. In this work, we provide a positive and constructive answer. We prove that a generic permutation-invariant function of $d$ objects can be asymptotically compressed into a function of $\operatorname{polylog} d$ objects with vanishing error, which is proved to be the optimal compression rate. This theorem yields two key implications: (Ia) a large neural network can be compressed to polylogarithmic width while preserving its learning dynamics; (Ib) a large dataset can be compressed to polylogarithmic size while leaving the loss landscape of the corresponding model unchanged. Implication (Ia) directly establishes a proof of the dynamical lottery ticket hypothesis, which states that any ordinary network can be strongly compressed such that the learning dynamics and result remain unchanged. (Ib) shows that a neural scaling law of the form $L\sim d^{-α}$ can be boosted to an arbitrarily fast power law decay, and ultimately to $\exp(-α' \sqrt(m){d})$.

---

### Diffusion Alignment as Variational Expectation-Maximization

Authors: Jaewoo Lee, Minsu Kim, Sanghyeok Choi, Inhyuck Song, Sujin Yun, Hyeongyu Kang, Woocheol Shin, Taeyoung Yun, Kiyoung Om, Jinkyoo Park
Published: 2025-10-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.00502v3

Diffusion alignment aims to optimize diffusion models for the downstream objective. While existing methods based on reinforcement learning or direct backpropagation achieve considerable success in maximizing rewards, they often suffer from reward over-optimization and mode collapse. We introduce Diffusion Alignment as Variational Expectation-Maximization (DAV), a framework that formulates diffusion alignment as an iterative process alternating between two complementary phases: the E-step and the M-step. In the E-step, we employ test-time search to generate diverse and reward-aligned samples. In the M-step, we refine the diffusion model using samples discovered by the E-step. We demonstrate that DAV can optimize reward while preserving diversity for both continuous and discrete tasks: text-to-image synthesis and DNA sequence design. Our code is available at https://github.com/Jaewoopudding/dav.

---

### Relative-Absolute Fusion: Rethinking Feature Extraction in Image-Based Iterative Method Selection for Solving Sparse Linear Systems

Authors: Kaiqi Zhang, Mingguan Yang, Dali Chang, Chun Chen, Yuxiang Zhang, Kexun He, Jing Zhao
Published: 2025-10-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.00500v1

Iterative method selection is crucial for solving sparse linear systems because these methods inherently lack robustness. Though image-based selection approaches have shown promise, their feature extraction techniques might encode distinct matrices into identical image representations, leading to the same selection and suboptimal method. In this paper, we introduce RAF (Relative-Absolute Fusion), an efficient feature extraction technique to enhance image-based selection approaches. By simultaneously extracting and fusing image representations as relative features with corresponding numerical values as absolute features, RAF achieves comprehensive matrix representations that prevent feature ambiguity across distinct matrices, thus improving selection accuracy and unlocking the potential of image-based selection approaches. We conducted comprehensive evaluations of RAF on SuiteSparse and our developed BMCMat (Balanced Multi-Classification Matrix dataset), demonstrating solution time reductions of 0.08s-0.29s for sparse linear systems, which is 5.86%-11.50% faster than conventional image-based selection approaches and achieves state-of-the-art (SOTA) performance. BMCMat is available at https://github.com/zkqq/BMCMat.

---

### MOSS-Speech: Towards True Speech-to-Speech Models Without Text Guidance

Authors: Xingjian Zhao, Zhe Xu, Qinyuan Cheng, Zhaoye Fei, Luozhijie Jin, Yang Wang, Hanfu Chen, Yaozhou Jiang, Qinghui Gao, Ke Chen, Ruixiao Li, Mingshu Chen et al.
Published: 2025-10-01
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.00499v2

Spoken dialogue systems often rely on cascaded pipelines that transcribe, process, and resynthesize speech. While effective, this design discards paralinguistic cues and limits expressivity. Recent end-to-end methods reduce latency and better preserve these cues, yet still rely on text intermediates, creating a fundamental bottleneck. We present MOSS-Speech, a true speech-to-speech large language model that directly understands and generates speech without relying on text guidance. Our approach combines a modality-based layer-splitting architecture with a frozen pre-training strategy, preserving the reasoning and knowledge of pretrained text LLMs while adding native speech capabilities. Experiments show that our model achieves state-of-the-art results in spoken question answering and delivers comparable speech-to-speech performance relative to existing text-guided systems, while still maintaining competitive text performance. By narrowing the gap between text-guided and direct speech generation, our work establishes a new paradigm for expressive and efficient end-to-end speech interaction.

---

### Emergence of robust looming selectivity via coordinated inhibitory neural computations

Authors: Qinbing Fu, Ziyan Qin
Published: 2025-10-01
Categories: q-bio.NC, cs.NE
arXiv: https://arxiv.org/abs/2510.00498v1

In the locust's lobula giant movement detector neural pathways, four categories of inhibition, i.e., global inhibition, self-inhibition, lateral inhibition, and feed-forward inhibition, have been functionally explored in the context of looming perception. However, their combined influence on shaping selectivity to looming motion remains unclear. Driven by recent physiological advancements, this paper offers new insights into the roles of these inhibitory mechanisms at multiple levels and scales in simulations, refining the specific selectivity for responding only to objects approaching the eyes while remaining unresponsive to other forms of movement. Within a feed-forward, multi-layer neural network framework, global inhibition, lateral inhibition, self-inhibition, and feed-forward inhibition are integrated. Global inhibition acts as an immediate feedback mechanism, normalising light intensities delivered by ommatidia, particularly addressing low-contrast looming. Self-inhibition, modelled numerically for the first time, suppresses translational motion. Lateral inhibition is formed by delayed local excitation spreading across a larger area. Notably, self-inhibition and lateral inhibition are sequential in time and are combined through feed-forward inhibition, which indicates the angular size subtended by moving objects. Together, these inhibitory processes attenuate motion-induced excitation at multiple levels and scales. This research suggests that self-inhibition may act earlier than lateral inhibition to rapidly reduce excitation in situ, thereby suppressing translational motion, and global inhibition can modulate excitation on a finer scale, enhancing selectivity in higher contrast range.

---

### Agent-ScanKit: Unraveling Memory and Reasoning of Multimodal Agents via Sensitivity Perturbations

Authors: Pengzhou Cheng, Lingzhong Dong, Zeng Wu, Zongru Wu, Xiangru Tang, Chengwei Qin, Zhuosheng Zhang, Gongshen Liu
Published: 2025-10-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.00496v3

Although numerous strategies have recently been proposed to enhance the autonomous interaction capabilities of multimodal agents in graphical user interface (GUI), their reliability remains limited when faced with complex or out-of-domain tasks. This raises a fundamental question: Are existing multimodal agents reasoning spuriously? In this paper, we propose \textbf{Agent-ScanKit}, a systematic probing framework to unravel the memory and reasoning capabilities of multimodal agents under controlled perturbations. Specifically, we introduce three orthogonal probing paradigms: visual-guided, text-guided, and structure-guided, each designed to quantify the contributions of memorization and reasoning without requiring access to model internals. In five publicly available GUI benchmarks involving 18 multimodal agents, the results demonstrate that mechanical memorization often outweighs systematic reasoning. Most of the models function predominantly as retrievers of training-aligned knowledge, exhibiting limited generalization. Our findings underscore the necessity of robust reasoning modeling for multimodal agents in real-world scenarios, offering valuable insights toward the development of reliable multimodal agents.

---

### Normal-Abnormal Guided Generalist Anomaly Detection

Authors: Yuexin Wang, Xiaolei Wang, Yizheng Gong, Jimin Xiao
Published: 2025-10-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.00495v3

Generalist Anomaly Detection (GAD) aims to train a unified model on an original domain that can detect anomalies in new target domains. Previous GAD methods primarily use only normal samples as references, overlooking the valuable information contained in anomalous samples that are often available in real-world scenarios. To address this limitation, we propose a more practical approach: normal-abnormal-guided generalist anomaly detection, which leverages both normal and anomalous samples as references to guide anomaly detection across diverse domains. We introduce the Normal-Abnormal Generalist Learning (NAGL) framework, consisting of two key components: Residual Mining (RM) and Anomaly Feature Learning (AFL). RM extracts abnormal patterns from normal-abnormal reference residuals to establish transferable anomaly representations, while AFL adaptively learns anomaly features in query images through residual mapping to identify instance-aware anomalies. Our approach effectively utilizes both normal and anomalous references for more accurate and efficient cross-domain anomaly detection. Extensive experiments across multiple benchmarks demonstrate that our method significantly outperforms existing GAD approaches. This work represents the first to adopt a mixture of normal and abnormal samples as references in generalist anomaly detection. The code and datasets are available at https://github.com/JasonKyng/NAGL.

---

### Exploring System 1 and 2 communication for latent reasoning in LLMs

Authors: Julian Coda-Forno, Zhuokai Zhao, Qiang Zhang, Dipesh Tamboli, Weiwei Li, Xiangjun Fan, Lizhu Zhang, Eric Schulz, Hsiao-Ping Tseng
Published: 2025-10-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.00494v2

Should LLM reasoning live in a separate module, or within a single model's forward pass and representational space? We study dual-architecture latent reasoning, where a fluent Base exchanges latent messages with a Coprocessor, and test two hypotheses aimed at improving latent communication over Liu et al. (2024): (H1) increase channel capacity; (H2) learn communication via joint finetuning. Under matched latent-token budgets on GPT-2 and Qwen-3, H2 is consistently strongest while H1 yields modest gains. A unified soft-embedding baseline, a single model with the same forward pass and shared representations, using the same latent-token budget, nearly matches H2 and surpasses H1, suggesting current dual designs mostly add compute rather than qualitatively improving reasoning. Across GSM8K, ProsQA, and a Countdown stress test with increasing branching factor, scaling the latent-token budget beyond small values fails to improve robustness. Latent analyses show overlapping subspaces with limited specialization, consistent with weak reasoning gains. We conclude dual-model latent reasoning remains promising in principle, but likely requires objectives and training schedules that explicitly shape latent spaces for algorithmic planning.

---

### Rethinking Reward Models for Multi-Domain Test-Time Scaling

Authors: Dong Bok Lee, Seanie Lee, Sangwoo Park, Minki Kang, Jinheon Baek, Dongki Kim, Dominik Wagner, Jiongdao Jin, Heejun Lee, Tobias Bocklet, Jinyu Wang, Jingjing Fu et al.
Published: 2025-10-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.00492v3

The reliability of large language models (LLMs) during test-time scaling is often assessed with \emph{external verifiers} or \emph{reward models} that distinguish correct reasoning from flawed logic. Prior work has studied both outcome reward models (ORMs), which assess only the final answer, and process reward models (PRMs), which score intermediate reasoning steps. Although PRMs are often viewed as advantageous due to their finer-grained supervision, much of the supporting evidence comes from math-adjacent settings, and their relative benefits across broader domains remain unclear. We present the first unified evaluation of four reward model variants, discriminative ORM and PRM (dORM, dPRM) and generative ORM and PRM (gORM, gPRM), across 14 diverse domains. Contrary to conventional wisdom, we find that (i) dORM performs on par with dPRM, (ii) gPRM is not competitive, and (iii) overall, gORM is the most robust, yielding significant and consistent gains across every tested domain. We attribute the worse performance of gPRM to the stepwise scoring process, which inherits label noise from LLM-based automatic labeling, leading to difficulties in evaluating long reasoning trajectories, including those involving self-correcting reasoning. Both our theoretical analysis and empirical observations indicate that stepwise aggregation compounds errors as reasoning length increases. These findings challenge the common assumption that fine-grained supervision is always better and support generative outcome verification for multi-domain deployment. Our \href{https://github.com/db-Lee/Multi-RM}{\underline{code}} is publicly available to facilitate future research in multi-domain settings.

---

### Traj2Action: A Co-Denoising Framework for Trajectory-Guided Human-to-Robot Skill Transfer

Authors: Han Zhou, Jinjin Cao, Liyuan Ma, Xueji Fang, Guo-jun Qi
Published: 2025-10-01
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.00491v3

Learning diverse manipulation skills for real-world robots is severely bottlenecked by the reliance on costly and hard-to-scale teleoperated demonstrations. While human videos offer a scalable alternative, effectively transferring manipulation knowledge is fundamentally hindered by the significant morphological gap between human and robotic embodiments. To address this challenge and facilitate skill transfer from human to robot, we introduce Traj2Action, a novel framework that bridges this embodiment gap by using the 3D trajectory of the operational endpoint as a unified intermediate representation, and then transfers the manipulation knowledge embedded in this trajectory to the robot's actions. Our policy first learns to generate a coarse trajectory, which forms a high-level motion plan by leveraging both human and robot data. This plan then conditions the synthesis of precise, robot-specific actions (e.g., orientation and gripper state) within a co-denoising framework. Our work centers on two core objectives: first, the systematic verification of the Traj2Action framework's effectiveness-spanning architectural design, cross-task generalization, and data efficiency and second, the revelation of key laws that govern robot policy learning during the integration of human hand demonstration data. This research focus enables us to provide a scalable paradigm tailored to address human-to-robot skill transfer across morphological gaps. Extensive real-world experiments on a Franka robot demonstrate that Traj2Action boosts the performance by up to 27% and 22.25% over $π_0$ baseline on short- and long-horizon real-world tasks, and achieves significant gains as human data scales in robot policy learning.

---

### Black-Box Time-Series Domain Adaptation via Cross-Prompt Foundation Models

Authors: M. T. Furqon, Mahardhika Pratama, Igor Skrjanc, Lin Liu, Habibullah Habibullah, Kutluyil Dogancay
Published: 2025-10-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.00487v1

The black-box domain adaptation (BBDA) topic is developed to address the privacy and security issues where only an application programming interface (API) of the source model is available for domain adaptations. Although the BBDA topic has attracted growing research attentions, existing works mostly target the vision applications and are not directly applicable to the time-series applications possessing unique spatio-temporal characteristics. In addition, none of existing approaches have explored the strength of foundation model for black box time-series domain adaptation (BBTSDA). This paper proposes a concept of Cross-Prompt Foundation Model (CPFM) for the BBTSDA problems. CPFM is constructed under a dual branch network structure where each branch is equipped with a unique prompt to capture different characteristics of data distributions. In the domain adaptation phase, the reconstruction learning phase in the prompt and input levels is developed. All of which are built upon a time-series foundation model to overcome the spatio-temporal dynamic. Our rigorous experiments substantiate the advantage of CPFM achieving improved results with noticeable margins from its competitors in three time-series datasets of different application domains.

---

### PodEval: A Multimodal Evaluation Framework for Podcast Audio Generation

Authors: Yujia Xiao, Liumeng Xue, Lei He, Xinyi Chen, Aemon Yat Fei Chiu, Wenjie Tian, Shaofei Zhang, Qiuqiang Kong, Xinfa Zhu, Wei Xue, Tan Lee
Published: 2025-10-01
Categories: cs.SD, cs.AI, eess.AS
arXiv: https://arxiv.org/abs/2510.00485v1

Recently, an increasing number of multimodal (text and audio) benchmarks have emerged, primarily focusing on evaluating models' understanding capability. However, exploration into assessing generative capabilities remains limited, especially for open-ended long-form content generation. Significant challenges lie in no reference standard answer, no unified evaluation metrics and uncontrollable human judgments. In this work, we take podcast-like audio generation as a starting point and propose PodEval, a comprehensive and well-designed open-source evaluation framework. In this framework: 1) We construct a real-world podcast dataset spanning diverse topics, serving as a reference for human-level creative quality. 2) We introduce a multimodal evaluation strategy and decompose the complex task into three dimensions: text, speech and audio, with different evaluation emphasis on "Content" and "Format". 3) For each modality, we design corresponding evaluation methods, involving both objective metrics and subjective listening test. We leverage representative podcast generation systems (including open-source, close-source, and human-made) in our experiments. The results offer in-depth analysis and insights into podcast generation, demonstrating the effectiveness of PodEval in evaluating open-ended long-form audio. This project is open-source to facilitate public use: https://github.com/yujxx/PodEval.

---

### MathSticks: A Benchmark for Visual Symbolic Compositional Reasoning with Matchstick Puzzles

Authors: Yuheng Ji, Huajie Tan, Cheng Chi, Yijie Xu, Yuting Zhao, Enshen Zhou, Huaihai Lyu, Pengwei Wang, Zhongyuan Wang, Shanghang Zhang, Xiaolong Zheng
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00483v1

We introduce \textsc{MathSticks}, a benchmark for Visual Symbolic Compositional Reasoning (VSCR), which unifies visual perception, symbolic manipulation, and arithmetic consistency. Each task presents an incorrect matchstick equation that must be corrected by moving one or two sticks under strict conservation rules. The benchmark includes both text-guided and purely visual settings, systematically covering digit scale, move complexity, solution multiplicity, and operator variation, with 1.4M generated instances and a curated test set. Evaluations of 14 vision--language models reveal substantial limitations: closed-source models succeed only on simple cases, open-source models fail in the visual regime, while humans exceed 90\% accuracy. These findings establish \textsc{MathSticks} as a rigorous testbed for advancing compositional reasoning across vision and symbols. Our code and dataset are publicly available at https://github.com/Yuheng2000/MathSticks.

---

### Agent Fine-tuning through Distillation for Domain-specific LLMs in Microdomains

Authors: Yawen Xue, Masaya Tsunokake, Yuta Koreeda, Ekant Muljibhai Amin, Takashi Sumiyoshi, Yasuhiro Sogawa
Published: 2025-10-01
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.00482v1

Agentic large language models (LLMs) have become prominent for autonomously interacting with external environments and performing multi-step reasoning tasks. Most approaches leverage these capabilities via in-context learning with few-shot prompts, but this often results in lengthy inputs and higher computational costs. Agent fine-tuning offers an alternative by enabling LLMs to internalize procedural reasoning and domain-specific knowledge through training on relevant data and demonstration trajectories. While prior studies have focused on general domains, their effectiveness in specialized technical microdomains remains unclear. This paper explores agent fine-tuning for domain adaptation within Hitachi's JP1 middleware, a microdomain for specialized IT operations. We fine-tuned LLMs using JP1-specific datasets derived from domain manuals and distilled reasoning trajectories generated by LLMs themselves, enhancing decision making accuracy and search efficiency. During inference, we used an agentic prompt with retrieval-augmented generation and introduced a context-answer extractor to improve information relevance. On JP1 certification exam questions, our method achieved a 14% performance improvement over the base model, demonstrating the potential of agent fine-tuning for domain-specific reasoning in complex microdomains.

---

### Make a Video Call with LLM: A Measurement Campaign over Six Mainstream Apps

Authors: Jiayang Xu, Xiangjie Huang, Zijie Li, Antariksh Verma, Zili Meng
Published: 2025-10-01
Categories: cs.NI, cs.AI, cs.HC, cs.MM, cs.PF
arXiv: https://arxiv.org/abs/2510.00481v2

In 2025, Large Language Model (LLM) services have launched a new feature -- AI video chat -- allowing users to interact with AI agents via real-time video communication (RTC), just like chatting with real people. Despite its significance, no systematic study has characterized the performance of existing AI video chat systems. To address this gap, this paper proposes a comprehensive benchmark across four dimensions: quality, latency, internal mechanisms, and system overhead. Using custom testbeds, we further evaluate six mainstream AI video chatbots with this benchmark. We also build an online platform for user study. The measurement leads to interesting findings that could be beneficial to the future optimizations. For example, the network latency of AI video chat matters not as much as human video chat. The capabilities of AI agents matters most in the user experience. Our benchmarking results also open up several research questions for future optimizations of AI video chatbots. Availability: https://callarena.net/ for the online evaluation platform and our open-sourced dataset and testbed.

---

### Expandable Decision-Making States for Multi-Agent Deep Reinforcement Learning in Soccer Tactical Analysis

Authors: Kenjiro Ide, Taiga Someya, Kohei Kawaguchi, Keisuke Fujii
Published: 2025-10-01
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.00480v2

Invasion team sports such as soccer produce a high-dimensional, strongly coupled state space as many players continuously interact on a shared field, challenging quantitative tactical analysis. Traditional rule-based analyses are intuitive, while modern predictive machine learning models often perform pattern-matching without explicit agent representations. The problem we address is how to build player-level agent models from data, whose learned values and policies are both tactically interpretable and robust across heterogeneous data sources. Here, we propose Expandable Decision-Making States (EDMS), a semantically enriched state representation that augments raw positions and velocities with relational variables (e.g., scoring of space, pass, and score), combined with an action-masking scheme that gives on-ball and off-ball agents distinct decision sets. Compared to prior work, EDMS maps learned value functions and action policies to human-interpretable tactical concepts (e.g., marking pressure, passing lanes, ball accessibility) instead of raw coordinate features, and aligns agent choices with the rules of play. In the experiments, EDMS with action masking consistently reduced both action-prediction loss and temporal-difference (TD) error compared to the baseline. Qualitative case studies and Q-value visualizations further indicate that EDMS highlights high-risk, high-reward tactical patterns (e.g., fast counterattacks and defensive breakthroughs). We also integrated our approach into an open-source library and demonstrated compatibility with multiple commercial and open datasets, enabling cross-provider evaluation and reproducible experiments.

---

### On the joint estimation of flow fields and particle properties from Lagrangian data

Authors: Ke Zhou, Samuel J. Grauer
Published: 2025-10-01
Categories: physics.flu-dyn, stat.ML
arXiv: https://arxiv.org/abs/2510.00479v3

We numerically investigate the feasibility and limits of jointly estimating flow fields and unknown particle properties (e.g., position, size, and density) from Lagrangian particle tracking (LPT) data. LPT offers time-resolved, volumetric measurements of particle trajectories, which are markers of the carrier fluid motion. However, experimental tracks are spatially sparse and potentially noisy, and the problem of reconstructing flow fields may be further complicated by inertial particle transport, such that particle slip velocities must be determined to access the velocity field of the carrier fluid. To address this problem, we develop a data assimilation framework that couples an Eulerian representation of the flow with Lagrangian particle models, enabling the simultaneous inference of carrier fields and particle properties under the governing equations of disperse multiphase flow. We show that flow fields and particle properties can be jointly estimated in three representative regimes: (1) In a turbulent boundary layer with noisy tracer tracks (St to 0), flow fields and true particle positions are jointly estimated, which amounts to a physics-informed particle tracking problem; (2) in homogeneous isotropic turbulence seeded with inertial particles (St ~ 1-5), we demonstrate simultaneous recovery of flow states and particle diameters, showing the feasibility of implicit particle characterization; and (3) in a compressible, shock-dominated flow, we report the first joint reconstructions of velocity, pressure, density, and inertial particle properties (diameter and density), highlighting both the potential and certain limits of joint estimation in supersonic regimes. A systematic sensitivity study reveals how the seeding density, noise level, and Stokes number govern reconstruction accuracy for our method.

---

### Vicinity-Guided Discriminative Latent Diffusion for Privacy-Preserving Domain Adaptation

Authors: Jing Wang, Wonho Bae, Jiahong Chen, Wenxu Wang, Junhyug Noh
Published: 2025-10-01
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.00478v3

Recent work on latent diffusion models (LDMs) has focused almost exclusively on generative tasks, leaving their potential for discriminative transfer largely unexplored. We introduce Discriminative Vicinity Diffusion (DVD), a novel LDM-based framework for a more practical variant of source-free domain adaptation (SFDA): the source provider may share not only a pre-trained classifier but also an auxiliary latent diffusion module, trained once on the source data and never exposing raw source samples. DVD encodes each source feature's label information into its latent vicinity by fitting a Gaussian prior over its k-nearest neighbors and training the diffusion network to drift noisy samples back to label-consistent representations. During adaptation, we sample from each target feature's latent vicinity, apply the frozen diffusion module to generate source-like cues, and use a simple InfoNCE loss to align the target encoder to these cues, explicitly transferring decision boundaries without source access. Across standard SFDA benchmarks, DVD outperforms state-of-the-art methods. We further show that the same latent diffusion module enhances the source classifier's accuracy on in-domain data and boosts performance in supervised classification and domain generalization experiments. DVD thus reinterprets LDMs as practical, privacy-preserving bridges for explicit knowledge transfer, addressing a core challenge in source-free domain adaptation that prior methods have yet to solve.

---

### Analyzing Latent Concepts in Code Language Models

Authors: Arushi Sharma, Vedant Pungliya, Christopher J. Quinn, Ali Jannesari
Published: 2025-10-01
Categories: cs.SE, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.00476v2

Interpreting the internal behavior of large language models trained on code remains a critical challenge, particularly for applications demanding trust, transparency, and semantic robustness. We propose Code Concept Analysis (CoCoA): a global post-hoc interpretability framework that uncovers emergent lexical, syntactic, and semantic structures in a code language model's representation space by clustering contextualized token embeddings into human-interpretable concept groups. We propose a hybrid annotation pipeline that combines static analysis tool-based syntactic alignment with prompt-engineered large language models (LLMs), enabling scalable labeling of latent concepts across abstraction levels. We analyse the distribution of concepts across layers and across three finetuning tasks. Emergent concept clusters can help identify unexpected latent interactions and be used to identify trends and biases within the model's learned representations. We further integrate LCA with local attribution methods to produce concept-grounded explanations, improving the coherence and interpretability of token-level saliency. Empirical evaluations across multiple models and tasks show that LCA discovers concepts that remain stable under semantic-preserving perturbations (average Cluster Sensitivity Index, CSI = 0.288) and evolve predictably with fine-tuning. In a user study on the programming-language classification task, concept-augmented explanations disambiguated token roles and improved human-centric explainability by 37 percentage points compared with token-level attributions using Integrated Gradients.

---

### Diagnosing Shortcut-Induced Rigidity in Continual Learning: The Einstellung Rigidity Index (ERI)

Authors: Kai Gu, Weishi Shi
Published: 2025-10-01
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2510.00475v1

Deep neural networks frequently exploit shortcut features, defined as incidental correlations between inputs and labels without causal meaning. Shortcut features undermine robustness and reduce reliability under distribution shifts. In continual learning (CL), the consequences of shortcut exploitation can persist and intensify: weights inherited from earlier tasks bias representation reuse toward whatever features most easily satisfied prior labels, mirroring the cognitive Einstellung effect, a phenomenon where past habits block optimal solutions. Whereas catastrophic forgetting erodes past skills, shortcut-induced rigidity throttles the acquisition of new ones. We introduce the Einstellung Rigidity Index (ERI), a compact diagnostic that disentangles genuine transfer from cue-inflated performance using three interpretable facets: (i) Adaptation Delay (AD), (ii) Performance Deficit (PD), and (iii) Relative Suboptimal Feature Reliance (SFR_rel). On a two-phase CIFAR-100 CL benchmark with a deliberately spurious magenta patch in Phase 2, we evaluate Naive fine-tuning (SGD), online Elastic Weight Consolidation (EWC_on), Dark Experience Replay (DER++), Gradient Projection Memory (GPM), and Deep Generative Replay (DGR). Across these continual learning methods, we observe that CL methods reach accuracy thresholds earlier than a Scratch-T2 baseline (negative AD) but achieve slightly lower final accuracy on patched shortcut classes (positive PD). Masking the patch improves accuracy for CL methods while slightly reducing Scratch-T2, yielding negative SFR_rel. This pattern indicates the patch acted as a distractor for CL models in this setting rather than a helpful shortcut.

---

### Feature Identification via the Empirical NTK

Authors: Jennifer Lin
Published: 2025-10-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.00468v4

We provide evidence that eigenanalysis of the empirical neural tangent kernel (eNTK) can surface feature directions in trained neural networks. Across three increasingly realistic settings -- a 1-layer MLP trained on modular addition, a 1-layer Transformer trained on modular addition and the pretrained language model Gemma-3-270M -- we show that top eigenspaces of the eNTK align with ground-truth or interpretable features. In the modular arithmetic examples, top eNTK eigenspaces align with the Fourier features used by the MLP and the Fourier features at seed-dependent frequencies used by the Transformer to implement known ground-truth algorithms. Moreover, the alignment of the relevant subspaces evolves over training, with its first derivative peaking near the onset of grokking. For Gemma-3-270M, we compute top eNTK eigendirections on a dataset of TinyStories context windows and check their alignment with an automatically-generated set of parts-of-speech and other grammatical feature directions. We find that the alignment of eNTK eigendirections with grammar features outperforms a same-budget baseline of PCA on model activations. These results suggest that eNTK eigenanalysis may provide a new handle towards identifying features in trained models for mechanistic interpretability.

---

### Rehearsal-free and Task-free Online Continual Learning With Contrastive Prompt

Authors: Aopeng Wang, Ke Deng, Yongli Ren, Jun Luo
Published: 2025-10-01
Categories: cs.LG, cs.CV
arXiv: https://arxiv.org/abs/2510.00467v1

The main challenge of continual learning is \textit{catastrophic forgetting}. Because of processing data in one pass, online continual learning (OCL) is one of the most difficult continual learning scenarios. To address catastrophic forgetting in OCL, some existing studies use a rehearsal buffer to store samples and replay them in the later learning process, other studies do not store samples but assume a sequence of learning tasks so that the task identities can be explored. However, storing samples may raise data security or privacy concerns and it is not always possible to identify the boundaries between learning tasks in one pass of data processing. It motivates us to investigate rehearsal-free and task-free OCL (F2OCL). By integrating prompt learning with an NCM classifier, this study has effectively tackled catastrophic forgetting without storing samples and without usage of task boundaries or identities. The extensive experimental results on two benchmarks have demonstrated the effectiveness of the proposed method.

---

### Integrating Offline Pre-Training with Online Fine-Tuning: A Reinforcement Learning Approach for Robot Social Navigation

Authors: Run Su, Hao Fu, Shuai Zhou, Yingao Fu
Published: 2025-10-01
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.00466v1

Offline reinforcement learning (RL) has emerged as a promising framework for addressing robot social navigation challenges. However, inherent uncertainties in pedestrian behavior and limited environmental interaction during training often lead to suboptimal exploration and distributional shifts between offline training and online deployment. To overcome these limitations, this paper proposes a novel offline-to-online fine-tuning RL algorithm for robot social navigation by integrating Return-to-Go (RTG) prediction into a causal Transformer architecture. Our algorithm features a spatiotem-poral fusion model designed to precisely estimate RTG values in real-time by jointly encoding temporal pedestrian motion patterns and spatial crowd dynamics. This RTG prediction framework mitigates distribution shift by aligning offline policy training with online environmental interactions. Furthermore, a hybrid offline-online experience sampling mechanism is built to stabilize policy updates during fine-tuning, ensuring balanced integration of pre-trained knowledge and real-time adaptation. Extensive experiments in simulated social navigation environments demonstrate that our method achieves a higher success rate and lower collision rate compared to state-of-the-art baselines. These results underscore the efficacy of our algorithm in enhancing navigation policy robustness and adaptability. This work paves the way for more reliable and adaptive robotic navigation systems in real-world applications.

---

### On the Adversarial Robustness of Learning-based Conformal Novelty Detection

Authors: Daofu Zhang, Mehrdad Pournaderi, Hanne M. Clifford, Yu Xiang, Pramod K. Varshney
Published: 2025-10-01
Categories: stat.ML, cs.LG, eess.SP, stat.ME
arXiv: https://arxiv.org/abs/2510.00463v4

This paper studies the adversarial robustness of conformal novelty detection. In particular, we focus on two powerful learning-based frameworks that come with finite-sample false discovery rate (FDR) control: one is AdaDetect (by Marandon et al., 2024) that is based on the positive-unlabeled classifier, and the other is a one-class classifier-based approach (by Bates et al., 2023). While they provide rigorous statistical guarantees under benign conditions, their behavior under adversarial perturbations remains underexplored. We first formulate an oracle attack setup, under the AdaDetect formulation, that quantifies the worst-case degradation of FDR, deriving an upper bound that characterizes the statistical cost of attacks. This idealized formulation directly motivates a practical and effective attack scheme that only requires query access to the output labels of both frameworks. Coupling these formulations with two popular and complementary black-box adversarial algorithms, we systematically evaluate the vulnerability of both frameworks on synthetic and real-world datasets. Our results show that adversarial perturbations can significantly increase the FDR while maintaining high detection power, exposing fundamental limitations of current error-controlled novelty detection methods and motivating the development of more robust alternatives.

---

### TimeEmb: A Lightweight Static-Dynamic Disentanglement Framework for Time Series Forecasting

Authors: Mingyuan Xia, Chunxu Zhang, Zijian Zhang, Hao Miao, Qidong Liu, Yuanshao Zhu, Bo Yang
Published: 2025-10-01
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.00461v2

Temporal non-stationarity, the phenomenon that time series distributions change over time, poses fundamental challenges to reliable time series forecasting. Intuitively, the complex time series can be decomposed into two factors, \ie time-invariant and time-varying components, which indicate static and dynamic patterns, respectively. Nonetheless, existing methods often conflate the time-varying and time-invariant components, and jointly learn the combined long-term patterns and short-term fluctuations, leading to suboptimal performance facing distribution shifts. To address this issue, we initiatively propose a lightweight static-dynamic decomposition framework, TimeEmb, for time series forecasting. TimeEmb innovatively separates time series into two complementary components: (1) time-invariant component, captured by a novel global embedding module that learns persistent representations across time series, and (2) time-varying component, processed by an efficient frequency-domain filtering mechanism inspired by full-spectrum analysis in signal processing. Experiments on real-world datasets demonstrate that TimeEmb outperforms state-of-the-art baselines and requires fewer computational resources. We conduct comprehensive quantitative and qualitative analyses to verify the efficacy of static-dynamic disentanglement. This lightweight framework can also improve existing time-series forecasting methods with simple integration. To ease reproducibility, the code is available at https://github.com/showmeon/TimeEmb.

---

### Robust Spatiotemporally Contiguous Anomaly Detection Using Tensor Decomposition

Authors: Rachita Mondal, Mert Indibi, Tapabrata Maiti, Selin Aviyente
Published: 2025-10-01
Categories: cs.LG, stat.ME, stat.ML
arXiv: https://arxiv.org/abs/2510.00460v1

Anomaly detection in spatiotemporal data is a challenging problem encountered in a variety of applications, including video surveillance, medical imaging data, and urban traffic monitoring. Existing anomaly detection methods focus mainly on point anomalies and cannot deal with temporal and spatial dependencies that arise in spatio-temporal data. Tensor-based anomaly detection methods have been proposed to address this problem. Although existing methods can capture dependencies across different modes, they are primarily supervised and do not account for the specific structure of anomalies. Moreover, these methods focus mainly on extracting anomalous features without providing any statistical confidence. In this paper, we introduce an unsupervised tensor-based anomaly detection method that simultaneously considers the sparse and spatiotemporally smooth nature of anomalies. The anomaly detection problem is formulated as a regularized robust low-rank + sparse tensor decomposition where the total variation of the tensor with respect to the underlying spatial and temporal graphs quantifies the spatiotemporal smoothness of the anomalies. Once the anomalous features are extracted, we introduce a statistical anomaly scoring framework that accounts for local spatio-temporal dependencies. The proposed framework is evaluated on both synthetic and real data.

---

### VLOD-TTA: Test-Time Adaptation of Vision-Language Object Detectors

Authors: Atif Belal, Heitor R. Medeiros, Marco Pedersoli, Eric Granger
Published: 2025-10-01
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.00458v4

Vision-language object detectors (VLODs) such as YOLO-World and Grounding DINO exhibit strong zero-shot generalization, but their performance degrades under distribution shift. Test-time adaptation (TTA) offers a practical way to adapt models during inference using only unlabeled target (test) data. However, while TTA has made substantial progress in vision-language classification, its application to VLODs remains largely unexplored. The only prior method relies on a mean-teacher framework that introduces significant latency and memory overhead. To this end, we introduce VLOD-TTA, a TTA method that leverages dense proposal overlap and image-conditioned prompts to adapt VLODs with low additional overhead. VLOD-TTA combines (i) an IoU-weighted entropy objective that emphasizes spatially coherent proposal clusters and mitigates confirmation bias from isolated boxes, and (ii) image-conditioned prompt selection that ranks prompts by image-level compatibility and aggregates the most informative prompt scores for detection. Our experiments across diverse distribution shifts, including artistic domains, adverse driving conditions, low-light imagery, and common corruptions, indicate that VLOD-TTA consistently outperforms standard TTA baselines and the prior state-of-the-art method using YOLO-World and Grounding DINO. Our code: https://github.com/imatif17/VLOD-TTA

---

### UrbanGraph: Physics-Informed Spatio-Temporal Dynamic Heterogeneous Graphs for Urban Microclimate Prediction

Authors: Weilin Xin, Chenyu Huang, Peilin Li, Jing Zhong, Jiawei Yao
Published: 2025-10-01
Categories: cs.LG, cs.AI, cs.CE
arXiv: https://arxiv.org/abs/2510.00457v2

With rapid urbanization, predicting urban microclimates has become critical, as it affects building energy demand and public health risks. However, existing generative and homogeneous graph approaches fall short in capturing physical consistency, spatial dependencies, and temporal variability. To address this, we introduce UrbanGraph, a framework founded on a novel structure-based inductive bias. Unlike implicit graph learning, UrbanGraph transforms physical first principles into a dynamic causal topology, explicitly encoding time-varying causalities (e.g., shading and convection) directly into the graph structure to ensure physical consistency and data efficiency. Results show that UrbanGraph achieves state-of-the-art performance across all baselines. Specifically, the use of explicit causal pruning significantly reduces the model's floating-point operations (FLOPs) by 73.8% and increases training speed by 21% compared to implicit graphs. Our contribution includes the first high-resolution benchmark for spatio-temporal microclimate modeling, and a generalizable explicit topological encoding paradigm applicable to urban spatio-temporal dynamics governed by known physical equations.

---

### Measuring and Controlling the Spectral Bias for Self-Supervised Image Denoising

Authors: Wang Zhang, Huaqiu Li, Xiaowan Hu, Tao Jiang, Zikang Chen, Haoqian Wang
Published: 2025-10-01
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.00454v1

Current self-supervised denoising methods for paired noisy images typically involve mapping one noisy image through the network to the other noisy image. However, after measuring the spectral bias of such methods using our proposed Image Pair Frequency-Band Similarity, it suffers from two practical limitations. Firstly, the high-frequency structural details in images are not preserved well enough. Secondly, during the process of fitting high frequencies, the network learns high-frequency noise from the mapped noisy images. To address these challenges, we introduce a Spectral Controlling network (SCNet) to optimize self-supervised denoising of paired noisy images. First, we propose a selection strategy to choose frequency band components for noisy images, to accelerate the convergence speed of training. Next, we present a parameter optimization method that restricts the learning ability of convolutional kernels to high-frequency noise using the Lipschitz constant, without changing the network structure. Finally, we introduce the Spectral Separation and low-rank Reconstruction module (SSR module), which separates noise and high-frequency details through frequency domain separation and low-rank space reconstruction, to retain the high-frequency structural details of images. Experiments performed on synthetic and real-world datasets verify the effectiveness of SCNet.

---