# Research Papers — 2025-10 · bundle page 182 of 217

> Full-text markdown of papers #9051–9100 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/182
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/181.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/183.md)

### ViTs: Teaching Machines to See Time Series Anomalies Like Human Experts

Authors: Zexin Wang, Changhua Pei, Yang Liu, Hengyue Jiang, Quan Zhou, Haotian Si, Hang Cui, Jianhui Li, Gaogang Xie, Jingjing Li, Dan Pei
Published: 2025-10-06
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.04710v1

Web service administrators must ensure the stability of multiple systems by promptly detecting anomalies in Key Performance Indicators (KPIs). Achieving the goal of "train once, infer across scenarios" remains a fundamental challenge for time series anomaly detection models. Beyond improving zero-shot generalization, such models must also flexibly handle sequences of varying lengths during inference, ranging from one hour to one week, without retraining. Conventional approaches rely on sliding-window encoding and self-supervised learning, which restrict inference to fixed-length inputs. Large Language Models (LLMs) have demonstrated remarkable zero-shot capabilities across general domains. However, when applied to time series data, they face inherent limitations due to context length. To address this issue, we propose ViTs, a Vision-Language Model (VLM)-based framework that converts time series curves into visual representations. By rescaling time series images, temporal dependencies are preserved while maintaining a consistent input size, thereby enabling efficient processing of arbitrarily long sequences without context constraints. Training VLMs for this purpose introduces unique challenges, primarily due to the scarcity of aligned time series image-text data. To overcome this, we employ an evolutionary algorithm to automatically generate thousands of high-quality image-text pairs and design a three-stage training pipeline consisting of: (1) time series knowledge injection, (2) anomaly detection enhancement, and (3) anomaly reasoning refinement. Extensive experiments demonstrate that ViTs substantially enhance the ability of VLMs to understand and detect anomalies in time series data. All datasets and code will be publicly released at: https://anonymous.4open.science/r/ViTs-C484/.

---

### ID-Consistent, Precise Expression Generation with Blendshape-Guided Diffusion

Authors: Foivos Paraperas Papantoniou, Stefanos Zafeiriou
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04706v1

Human-centric generative models designed for AI-driven storytelling must bring together two core capabilities: identity consistency and precise control over human performance. While recent diffusion-based approaches have made significant progress in maintaining facial identity, achieving fine-grained expression control without compromising identity remains challenging. In this work, we present a diffusion-based framework that faithfully reimagines any subject under any particular facial expression. Building on an ID-consistent face foundation model, we adopt a compositional design featuring an expression cross-attention module guided by FLAME blendshape parameters for explicit control. Trained on a diverse mixture of image and video data rich in expressive variation, our adapter generalizes beyond basic emotions to subtle micro-expressions and expressive transitions, overlooked by prior works. In addition, a pluggable Reference Adapter enables expression editing in real images by transferring the appearance from a reference frame during synthesis. Extensive quantitative and qualitative evaluations show that our model outperforms existing methods in tailored and identity-consistent expression generation. Code and models can be found at https://github.com/foivospar/Arc2Face.

---

### Label-Efficient Cross-Modality Generalization for Liver Segmentation in Multi-Phase MRI

Authors: Quang-Khai Bui-Tran, Minh-Toan Dinh, Thanh-Huy Nguyen, Ba-Thinh Lam, Mai-Anh Vu, Ulas Bagci
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04705v4

Accurate liver segmentation in multi-phase MRI is vital for liver fibrosis assessment, yet labeled data is often scarce and unevenly distributed across imaging modalities and vendor systems. We propose a label-efficient segmentation approach that promotes cross-modality generalization under real-world conditions, where GED4 hepatobiliary-phase annotations are limited, non-contrast sequences (T1WI, T2WI, DWI) are unlabeled, and spatial misalignment and missing phases are common. Our method integrates a foundation-scale 3D segmentation backbone adapted via fine-tuning, co-training with cross pseudo supervision to leverage unlabeled volumes, and a standardized preprocessing pipeline. Without requiring spatial registration, the model learns to generalize across MRI phases and vendors, demonstrating robust segmentation performance in both labeled and unlabeled domains. Our results exhibit the effectiveness of our proposed label-efficient baseline for liver segmentation in multi-phase, multi-vendor MRI and highlight the potential of combining foundation model adaptation with co-training for real-world clinical imaging tasks.

---

### AtomWorld: A Benchmark for Evaluating Spatial Reasoning in Large Language Models on Crystalline Materials

Authors: Taoyuze Lv, Alexander Chen, Fengyu Xie, Chu Wu, Jeffrey Meng, Dongzhan Zhou, Yingheng Wang, Bram Hoex, Zhicheng Zhong, Tong Xie
Published: 2025-10-06
Categories: cond-mat.mtrl-sci, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.04704v4

Large language models (LLMs) have shown promising potential in scientific research, enabling tasks ranging from knowledge retrieval to property prediction. Existing science benchmarks mainly focus on perceptual or knowledge-based tasks, largely ignoring the modelling tasks, a fundamental starting point for any real scientific research. For materials science, constructing and manipulating atomic structures is one of the most creative and least automated steps. In this work, we introduce AtomWorld, a benchmark designed to evaluate the abilities of LLMs on structure modifications. The benchmark includes ten fundamental actions under four widely used modelling categories, enabling verifiable evaluation metrics. We find that Claude Opus 4.6 generally performs the best. While the success rate decreases markedly with increasing modelling complexity, with particularly low success rates (below 12\% for rotation) for operations involving complex spatial relations. Our results suggest that contemporary LLMs are better suited as copilots for materials structure modelling rather than fully unsupervised autonomous scientific agents. Beyond evaluation, AtomWorld also serves as a testbed and playground for developing future structure-aware models, including reinforcement learning and agentic approaches.

---

### The Bayesian Origin of the Probability Weighting Function in Human Representation of Probabilities

Authors: Xin Tong, Thi Thu Uyen Hoang, Xue-Xin Wei, Michael Hahn
Published: 2025-10-06
Categories: q-bio.NC, cs.AI, econ.TH
arXiv: https://arxiv.org/abs/2510.04698v3

Humans systematically misrepresent probability in a stereotyped inverse-S pattern. It has been documented for decades, but its origin remains unexplained. We propose a Bayesian encoding-decoding account in which probabilities are represented by noisy internal signals and decoded by Bayes-risk minimization. For bounded probability stimuli, we show that distortion decomposes into boundary regression, likelihood repulsion, and prior attraction, yielding a key prediction: the classic inverse-S-shaped weighting pattern implies a U-shaped allocation of encoding precision with greater sensitivity near 0 and 1. Across judgment of relative frequency, lottery pricing, and risky choice, this U-shape is recovered from data without imposing any functional form on the encoding, and our framework outperforms deterministic weighting functions, bounded log-odds models, uniform-encoding Bayesian accounts, and matched efficient-coding models on held-out data. In a new dot probability estimation experiment with bimodal stimulus statistics, the recovered prior tracks the new distribution while the recovered encoding remains U-shaped. Together, these results identify the inverse-S-shaped probability weighting function as the joint product of a stable U-shaped encoding and a flexible prior, integrated by optimal Bayesian decoding.

---

### Beyond Outcome Reward: Decoupling Search and Answering Improves LLM Agents

Authors: Yiding Wang, Zhepei Wei, Xinyu Zhu, Yu Meng
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04695v1

Enabling large language models (LLMs) to utilize search tools offers a promising path to overcoming fundamental limitations such as knowledge cutoffs and hallucinations. Recent work has explored reinforcement learning (RL) for training search-augmented agents that interleave reasoning and retrieval before answering. These approaches usually rely on outcome-based rewards (e.g., exact match), implicitly assuming that optimizing for final answers will also yield effective intermediate search behaviors. Our analysis challenges this assumption: we uncover multiple systematic deficiencies in search that arise under outcome-only training and ultimately degrade final answer quality, including failure to invoke tools, invalid queries, and redundant searches. To address these shortcomings, we introduce DeSA (Decoupling Search-and-Answering), a simple two-stage training framework that explicitly separates search optimization from answer generation. In Stage 1, agents are trained to improve search effectiveness with retrieval recall-based rewards. In Stage 2, outcome rewards are employed to optimize final answer generation. Across seven QA benchmarks, DeSA-trained agents consistently improve search behaviors, delivering substantially higher search recall and answer accuracy than outcome-only baselines. Notably, DeSA outperforms single-stage training approaches that simultaneously optimize recall and outcome rewards, underscoring the necessity of explicitly decoupling the two objectives.

---

### Multilingual Routing in Mixture-of-Experts

Authors: Lucas Bandarkar, Chenyuan Yang, Mohsen Fayyaz, Junlin Hu, Nanyun Peng
Published: 2025-10-06
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.04694v2

Mixture-of-Experts (MoE) architectures have become the key to scaling modern LLMs, yet little is understood about how their sparse routing dynamics respond to multilingual data. In this work, we analyze expert routing patterns using parallel multilingual datasets and present highly interpretable layer-wise phenomena. We find that MoE models route tokens in language-specific ways in the early and late decoder layers but exhibit significant cross-lingual routing alignment in middle layers, mirroring parameter-sharing trends observed in dense LLMs. In particular, we reveal a clear, strong correlation between a model's performance in a given language and how similarly its tokens are routed to English in these layers. Extending beyond correlation, we explore inference-time interventions that induce higher cross-lingual routing alignment. We introduce a method that steers the router by promoting middle-layer task experts frequently activated in English, and it successfully increases multilingual performance. These 1-2% gains are remarkably consistent across two evaluation tasks, three models, and 15+ languages, especially given that these simple interventions override routers of extensively trained, state-of-the-art LLMs. In comparison, interventions outside of the middle layers or targeting multilingual-specialized experts only yield performance degradation. Altogether, we present numerous findings that explain how MoEs process non-English text and demonstrate that generalization is limited by the model's ability to leverage language-universal experts in all languages.

---

### Bio-Inspired Robotic Houbara: From Development to Field Deployment for Behavioral Studies

Authors: Lyes Saad Saoud, Irfan Hussain
Published: 2025-10-06
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2510.04692v1

Biomimetic intelligence and robotics are transforming field ecology by enabling lifelike robotic surrogates that interact naturally with animals under real world conditions. Studying avian behavior in the wild remains challenging due to the need for highly realistic morphology, durable outdoor operation, and intelligent perception that can adapt to uncontrolled environments. We present a next generation bio inspired robotic platform that replicates the morphology and visual appearance of the female Houbara bustard to support controlled ethological studies and conservation oriented field research. The system introduces a fully digitally replicable fabrication workflow that combines high resolution structured light 3D scanning, parametric CAD modelling, articulated 3D printing, and photorealistic UV textured vinyl finishing to achieve anatomically accurate and durable robotic surrogates. A six wheeled rocker bogie chassis ensures stable mobility on sand and irregular terrain, while an embedded NVIDIA Jetson module enables real time RGB and thermal perception, lightweight YOLO based detection, and an autonomous visual servoing loop that aligns the robot's head toward detected targets without human intervention. A lightweight thermal visible fusion module enhances perception in low light conditions. Field trials in desert aviaries demonstrated reliable real time operation at 15 to 22 FPS with latency under 100 ms and confirmed that the platform elicits natural recognition and interactive responses from live Houbara bustards under harsh outdoor conditions. This integrated framework advances biomimetic field robotics by uniting reproducible digital fabrication, embodied visual intelligence, and ecological validation, providing a transferable blueprint for animal robot interaction research, conservation robotics, and public engagement.

---

### A Study on the Data Distribution Gap in Music Emotion Recognition

Authors: Joann Ching, Gerhard Widmer
Published: 2025-10-06
Categories: cs.SD, cs.LG
arXiv: https://arxiv.org/abs/2510.04688v1

Music Emotion Recognition (MER) is a task deeply connected to human perception, relying heavily on subjective annotations collected from contributors. Prior studies tend to focus on specific musical styles rather than incorporating a diverse range of genres, such as rock and classical, within a single framework. In this paper, we address the task of recognizing emotion from audio content by investigating five datasets with dimensional emotion annotations -- EmoMusic, DEAM, PMEmo, WTC, and WCMED -- which span various musical styles. We demonstrate the problem of out-of-distribution generalization in a systematic experiment. By closely looking at multiple data and feature sets, we provide insight into genre-emotion relationships in existing data and examine potential genre dominance and dataset biases in certain feature representations. Based on these experiments, we arrive at a simple yet effective framework that combines embeddings extracted from the Jukebox model with chroma features and demonstrate how, alongside a combination of several diverse training sets, this permits us to train models with substantially improved cross-dataset generalization capabilities.

---

### How does the optimizer implicitly bias the model merging loss landscape?

Authors: Chenxiang Zhang, Alexander Theus, Damien Teney, Antonio Orvieto, Jun Pang, Sjouke Mauw
Published: 2025-10-06
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.04686v2

Model merging combines independent solutions with different capabilities into a single one while maintaining the same inference cost. Two popular approaches are linear interpolation, which simply averages multiple model weights, and task arithmetic, which combines task vectors obtained by the difference between finetuned and base models. While useful in practice, what properties make merging effective are poorly understood. This paper explores how the optimization dynamics affect the loss landscape geometry and its impact on merging success. We show that a single quantity -- the effective noise scale -- unifies the impact of different optimizer components on model merging. Across architectures and datasets, merging success is a non-monotonic function of the effective noise scale, with a distinct optimum. Decomposing this quantity, we find that larger learning rates, stronger weight decay, smaller batch sizes, and data augmentation all independently modulate the effective noise scale and exhibit the same qualitative trend. Unlike prior work connecting optimizer noise to the flatness or generalization of individual minima, we show that it also affects the global loss landscape, predicting when independently trained solutions can be successfully merged. Our findings broaden the understanding of how optimization shapes the loss landscape geometry and its consequences for model merging, suggesting that training dynamics could be further manipulated to improve model merging.

---

### Parameter-free Algorithms for the Stochastically Extended Adversarial Model

Authors: Shuche Wang, Adarsh Barik, Peng Zhao, Vincent Y. F. Tan
Published: 2025-10-06
Categories: cs.LG, math.OC
arXiv: https://arxiv.org/abs/2510.04685v1

We develop the first parameter-free algorithms for the Stochastically Extended Adversarial (SEA) model, a framework that bridges adversarial and stochastic online convex optimization. Existing approaches for the SEA model require prior knowledge of problem-specific parameters, such as the diameter of the domain $D$ and the Lipschitz constant of the loss functions $G$, which limits their practical applicability. Addressing this, we develop parameter-free methods by leveraging the Optimistic Online Newton Step (OONS) algorithm to eliminate the need for these parameters. We first establish a comparator-adaptive algorithm for the scenario with unknown domain diameter but known Lipschitz constant, achieving an expected regret bound of $\tilde{O}\big(\|u\|_2^2 + \|u\|_2(\sqrt{σ^2_{1:T}} + \sqrt{Σ^2_{1:T}})\big)$, where $u$ is the comparator vector and $σ^2_{1:T}$ and $Σ^2_{1:T}$ represent the cumulative stochastic variance and cumulative adversarial variation, respectively. We then extend this to the more general setting where both $D$ and $G$ are unknown, attaining the comparator- and Lipschitz-adaptive algorithm. Notably, the regret bound exhibits the same dependence on $σ^2_{1:T}$ and $Σ^2_{1:T}$, demonstrating the efficacy of our proposed methods even when both parameters are unknown in the SEA model.

---

### TiTok: Transfer Token-level Knowledge via Contrastive Excess to Transplant LoRA

Authors: Chanjoo Jung, Jaehyung Kim
Published: 2025-10-06
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.04682v3

Large Language Models (LLMs) are widely applied in real world scenarios, yet fine-tuning them comes with significant computational and storage costs. Parameter-Efficient Fine-Tuning (PEFT) methods such as LoRA mitigate these costs; however, the adapted parameters are dependent on the base model and cannot be transferred across different backbones. One way to address this issue is through knowledge distillation, but its effectiveness inherently depends on training data. Recent work such as TransLoRA avoids this by generating synthetic data; nevertheless, this adds complexity since it requires training an additional discriminator model. In this paper, we propose TiTok, a new framework that enables effective LoRA Transplantation through Token-level knowledge transfer. Specifically, TiTok captures task-relevant information through a token-wise contrastive excess between a source model with and without LoRA. This excess highlights informative tokens and enables selective filtering of synthetic data, all without additional models or overhead. Through experiments on three benchmarks across multiple transfer settings, we demonstrate that TiTok is consistently effective, achieving average performance gains of +4~10% compared to baselines overall.

---

### Multi-Agent Tool-Integrated Policy Optimization

Authors: Zhanfeng Mo, Xingxuan Li, Yuntao Chen, Lidong Bing
Published: 2025-10-06
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.04678v1

Large language models (LLMs) increasingly rely on multi-turn tool-integrated planning for knowledge-intensive and complex reasoning tasks. Existing implementations typically rely on a single agent, but they suffer from limited context length and noisy tool responses. A natural solution is to adopt a multi-agent framework with planner- and worker-agents to manage context. However, no existing methods support effective reinforcement learning post-training of tool-integrated multi-agent frameworks. To address this gap, we propose Multi-Agent Tool-Integrated Policy Optimization (MATPO), which enables distinct roles (planner and worker) to be trained within a single LLM instance using role-specific prompts via reinforcement learning. MATPO is derived from a principled credit assignment mechanism across planner and worker rollouts. This design eliminates the need to deploy multiple LLMs, which would be memory-intensive, while preserving the benefits of specialization. Experiments on GAIA-text, WebWalkerQA, and FRAMES show that MATPO consistently outperforms single-agent baselines by an average of 18.38% relative improvement in performance and exhibits greater robustness to noisy tool outputs. Our findings highlight the effectiveness of unifying multiple agent roles within a single LLM and provide practical insights for stable and efficient multi-agent RL training.

---

### Counterfactual Credit Guided Bayesian Optimization

Authors: Qiyu Wei, Haowei Wang, Richard Allmendinger, Mauricio A. Álvarez
Published: 2025-10-06
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.04676v2

Bayesian optimization has emerged as a prominent methodology for optimizing expensive black-box functions by leveraging Gaussian process surrogates, which focus on capturing the global characteristics of the objective function. However, in numerous practical scenarios, the primary objective is not to construct an exhaustive global surrogate, but rather to quickly pinpoint the global optimum. Due to the aleatoric nature of the sequential optimization problem and its dependence on the quality of the surrogate model and the initial design, it is restrictive to assume that all observed samples contribute equally to the discovery of the optimum in this context. In this paper, we introduce Counterfactual Credit Guided Bayesian Optimization (CCGBO), a novel framework that explicitly quantifies the contribution of individual historical observations through counterfactual credit. By incorporating counterfactual credit into the acquisition function, our approach can selectively allocate resources in areas where optimal solutions are most likely to occur. We prove that CCGBO retains sublinear regret. Empirical evaluations on various synthetic and real-world benchmarks demonstrate that CCGBO consistently reduces simple regret and accelerates convergence to the global optimum.

---

### Semantic Channel Equalization Strategies for Deep Joint Source-Channel Coding

Authors: Lorenzo Pannacci, Simone Fiorellino, Mario Edoardo Pandolfo, Emilio Calvanese Strinati, Paolo Di Lorenzo
Published: 2025-10-06
Categories: cs.LG, cs.AI, cs.IT, cs.NI
arXiv: https://arxiv.org/abs/2510.04674v1

Deep joint source-channel coding (DeepJSCC) has emerged as a powerful paradigm for end-to-end semantic communications, jointly learning to compress and protect task-relevant features over noisy channels. However, existing DeepJSCC schemes assume a shared latent space at transmitter (TX) and receiver (RX) - an assumption that fails in multi-vendor deployments where encoders and decoders cannot be co-trained. This mismatch introduces "semantic noise", degrading reconstruction quality and downstream task performance. In this paper, we systematize and evaluate methods for semantic channel equalization for DeepJSCC, introducing an additional processing stage that aligns heterogeneous latent spaces under both physical and semantic impairments. We investigate three classes of aligners: (i) linear maps, which admit closed-form solutions; (ii) lightweight neural networks, offering greater expressiveness; and (iii) a Parseval-frame equalizer, which operates in zero-shot mode without the need for training. Through extensive experiments on image reconstruction over AWGN and fading channels, we quantify trade-offs among complexity, data efficiency, and fidelity, providing guidelines for deploying DeepJSCC in heterogeneous AI-native wireless networks.

---

### Watch and Learn: Learning to Use Computers from Online Videos

Authors: Chan Hee Song, Yiwen Song, Palash Goyal, Yu Su, Oriana Riva, Hamid Palangi, Tomas Pfister
Published: 2025-10-06
Categories: cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2510.04673v3

Computer-using agents (CUAs) must plan task workflows across diverse and evolving applications, yet progress is limited by the lack of large-scale, high-quality training data. Existing datasets are narrow, static, and costly to annotate, while synthetic data often yields oversimplified or misaligned behaviors. We present Watch & Learn (W&L), a framework that converts readily available Internet videos of human computer use into executable UI trajectories at scale. Instead of directly generating actions or relying on handcrafted heuristics, we cast trajectory annotation as an inverse dynamics problem that predicts user actions from consecutive screen states, which simplifies learning and generalizes across domains. Through a task-aware retrieval and labeling pipeline, W&L yields over 53K high-quality trajectories that enhance CUAs both as in-context exemplars and as supervised training data. On OSWorld, it consistently improves general-purpose and specialized CUAs, while on WindowsAgentArena it achieves state-of-the-art performance among 7B-scale models under the 15-step limit. These results show that web-scale human demonstration videos can serve as a practical and scalable foundation for advancing real-world CUAs.

---

### FocusMed: A Large Language Model-based Framework for Enhancing Medical Question Summarization with Focus Identification

Authors: Chao Liu, Ling Luo, Tengxiao Lv, Huan Zhuang, Lejing Yu, Jian Wang, Hongfei Lin
Published: 2025-10-06
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.04671v1

With the rapid development of online medical platforms, consumer health questions (CHQs) are inefficient in diagnosis due to redundant information and frequent non-professional terms. The medical question summary (MQS) task aims to transform CHQs into streamlined doctors' frequently asked questions (FAQs), but existing methods still face challenges such as poor identification of question focus and model hallucination. This paper explores the potential of large language models (LLMs) in the MQS task and finds that direct fine-tuning is prone to focus identification bias and generates unfaithful content. To this end, we propose an optimization framework based on core focus guidance. First, a prompt template is designed to drive the LLMs to extract the core focus from the CHQs that is faithful to the original text. Then, a fine-tuning dataset is constructed in combination with the original CHQ-FAQ pairs to improve the ability to identify the focus of the question. Finally, a multi-dimensional quality evaluation and selection mechanism is proposed to comprehensively improve the quality of the summary from multiple dimensions. We conduct comprehensive experiments on two widely-adopted MQS datasets using three established evaluation metrics. The proposed framework achieves state-of-the-art performance across all measures, demonstrating a significant boost in the model's ability to identify critical focus of questions and a notable mitigation of hallucinations. The source codes are freely available at https://github.com/DUT-LiuChao/FocusMed.

---

### Improving Multimodal Brain Encoding Model with Dynamic Subject-awareness Routing

Authors: Xuanhua Yin, Runkai Zhao, Weidong Cai
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04670v3

Naturalistic fMRI encoding must handle multimodal inputs, shifting fusion styles, and pronounced inter-subject variability. We introduce AFIRE (Agnostic Framework for Multimodal fMRI Response Encoding), an agnostic interface that standardizes time-aligned post-fusion tokens from varied encoders, and MIND, a plug-and-play Mixture-of-Experts decoder with a subject-aware dynamic gating. Trained end-to-end for whole-brain prediction, AFIRE decouples the decoder from upstream fusion, while MIND combines token-dependent Top-K sparse routing with a subject prior to personalize expert usage without sacrificing generality. Experiments across multiple multimodal backbones and subjects show consistent improvements over strong baselines, enhanced cross-subject generalization, and interpretable expert patterns that correlate with content type. The framework offers a simple attachment point for new encoders and datasets, enabling robust, plug-and-improve performance for naturalistic neuroimaging studies.

---

### ConceptSplit: Decoupled Multi-Concept Personalization of Diffusion Models via Token-wise Adaptation and Attention Disentanglement

Authors: Habin Lim, Yeongseob Won, Juwon Seo, Gyeong-Moon Park
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04668v1

In recent years, multi-concept personalization for text-to-image (T2I) diffusion models to represent several subjects in an image has gained much more attention. The main challenge of this task is "concept mixing", where multiple learned concepts interfere or blend undesirably in the output image. To address this issue, in this paper, we present ConceptSplit, a novel framework to split the individual concepts through training and inference. Our framework comprises two key components. First, we introduce Token-wise Value Adaptation (ToVA), a merging-free training method that focuses exclusively on adapting the value projection in cross-attention. Based on our empirical analysis, we found that modifying the key projection, a common approach in existing methods, can disrupt the attention mechanism and lead to concept mixing. Second, we propose Latent Optimization for Disentangled Attention (LODA), which alleviates attention entanglement during inference by optimizing the input latent. Through extensive qualitative and quantitative experiments, we demonstrate that ConceptSplit achieves robust multi-concept personalization, mitigating unintended concept interference. Code is available at https://github.com/KU-VGI/ConceptSplit

---

### Noise or Signal? Deconstructing Contradictions and An Adaptive Remedy for Reversible Normalization in Time Series Forecasting

Authors: Fanzhe Fu, Yang Yang
Published: 2025-10-06
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.04667v1

Reversible Instance Normalization (RevIN) is a key technique enabling simple linear models to achieve state-of-the-art performance in time series forecasting. While replacing its non-robust statistics with robust counterparts (termed R$^2$-IN) seems like a straightforward improvement, our findings reveal a far more complex reality. This paper deconstructs the perplexing performance of various normalization strategies by identifying four underlying theoretical contradictions. Our experiments provide two crucial findings: first, the standard RevIN catastrophically fails on datasets with extreme outliers, where its MSE surges by a staggering 683\%. Second, while the simple R$^2$-IN prevents this failure and unexpectedly emerges as the best overall performer, our adaptive model (A-IN), designed to test a diagnostics-driven heuristic, unexpectedly suffers a complete and systemic failure. This surprising outcome uncovers a critical, overlooked pitfall in time series analysis: the instability introduced by a simple or counter-intuitive heuristic can be more damaging than the statistical issues it aims to solve. The core contribution of this work is thus a new, cautionary paradigm for time series normalization: a shift from a blind search for complexity to a diagnostics-driven analysis that reveals not only the surprising power of simple baselines but also the perilous nature of naive adaptation.

---

### An Attention-based Feature Memory Design for Energy-Efficient Continual Learning

Authors: Yuandou Wang, Filip Gunnarsson, Rihan Hai
Published: 2025-10-06
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.04660v2

Tabular data streams are increasingly prevalent in real-time decision-making across healthcare, finance, and the Internet of Things, often generated and processed on resource-constrained edge and mobile devices. Continual learning (CL) enables models to learn sequentially from such streams while retaining previously acquired knowledge. While recent CL advances have made significant progress in mitigating catastrophic forgetting, the energy and memory efficiency of CL for tabular data streams remains largely unexplored. To address this gap, we propose AttenMLP, which integrates attention-based feature replay with context retrieval and sliding buffer updates within a minibatch training framework for streaming tabular learning. We evaluate AttenMLP against state-of-the-art (SOTA) tabular models on real-world concept drift benchmarks with temporal distribution shifts. Experimental results show that AttenMLP achieves accuracy comparable to strong baselines without replay, while substantially reducing energy consumption through tunable design choices. In particular, with the proposed attention-based feature memory design, AttenMLP costs a 0.062 decrease in final accuracy under the incremental concept drift dataset, while reducing energy usage up to 33.3\% compared to TabPFNv2. Under the abrupt concept drift dataset, AttenMLP reduces 1.47\% energy consumption compared to TabR, at the cost of a 0.038 decrease in final accuracy. Although ranking third in global efficiency, AttenMLP demonstrates energy-accuracy trade-offs across both abrupt and incremental concept drift scenarios compared to SOTA tabular models.

---

### FT-MDT: Extracting Decision Trees from Medical Texts via a Novel Low-rank Adaptation Method

Authors: Yuheng Li, Jiechao Gao, Wei Han, Wenwen Ouyang, Wei Zhu, Hui Yi Leong
Published: 2025-10-06
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.04655v2

Knowledge of the medical decision process, which can be modeled as medical decision trees (MDTs), is critical to building clinical decision support systems. However, current MDT construction methods rely heavily on time-consuming and laborious manual annotation. To address this challenge, we propose PI-LoRA (Path-Integrated LoRA), a novel low-rank adaptation method for automatically extracting MDTs from clinical guidelines and textbooks. We integrate gradient path information to capture synergistic effects between different modules, enabling more effective and reliable rank allocation. This framework ensures that the most critical modules receive appropriate rank allocations while less important ones are pruned, resulting in a more efficient and accurate model for extracting medical decision trees from clinical texts. Extensive experiments on medical guideline datasets demonstrate that our PI-LoRA method significantly outperforms existing parameter-efficient fine-tuning approaches for the Text2MDT task, achieving better accuracy with substantially reduced model complexity. The proposed method achieves state-of-the-art results while maintaining a lightweight architecture, making it particularly suitable for clinical decision support systems where computational resources may be limited.

---

### MoME: Estimating Psychological Traits from Gait with Multi-Stage Mixture of Movement Experts

Authors: Andy Cǎtrunǎ, Adrian Cosma, Emilian Rǎdoi
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04654v2

Gait encodes rich biometric and behavioural information, yet leveraging the manner of walking to infer psychological traits remains a challenging and underexplored problem. We introduce a hierarchical Multi-Stage Mixture of Movement Experts (MoME) architecture for multi-task prediction of psychological attributes from gait sequences represented as 2D poses. MoME processes the walking cycle in four stages of movement complexity, employing lightweight expert models to extract spatio-temporal features and task-specific gating modules to adaptively weight experts across traits and stages. Evaluated on the PsyMo benchmark covering 17 psychological traits, our method outperforms state-of-the-art gait analysis models, achieving a 37.47% weighted F1 score at the run level and 44.6% at the subject level. Our experiments show that integrating auxiliary tasks such as identity recognition, gender prediction, and BMI estimation further improves psychological trait estimation. Our findings demonstrate the viability of multi-task gait-based learning for psychological trait estimation and provide a foundation for future research on movement-informed psychological inference.

---

### EduPersona: Benchmarking Subjective Ability Boundaries of Virtual Student Agents

Authors: Buyuan Zhu, Shiyu Hu, Yiping Ma, Yuanming Zhang, Kang Hao Cheong
Published: 2025-10-06
Categories: cs.CV, cs.CY
arXiv: https://arxiv.org/abs/2510.04648v1

As large language models are increasingly integrated into education, virtual student agents are becoming vital for classroom simulation and teacher training. Yet their classroom-oriented subjective abilities remain largely unassessed, limiting understanding of model boundaries and hindering trustworthy deployment. We present EduPersona, a large-scale benchmark spanning two languages, three subjects, and ten persona types based on the Big Five theory. The dataset contains 1,308 authentic classroom dialogue rounds, corresponding to 12,814 teacher-student Q&A turns, and is further expanded through persona stylization into roughly 10 times larger scale (128k turns), providing a solid foundation for evaluation. Building on this resource, we decompose hard-to-quantify subjective performance into three progressive tasks: TASK1 basic coherence (whether behavior, emotion, expression, and voice align with classroom context), TASK2 student realism, and TASK3 long-term persona consistency, thereby establishing an evaluation framework grounded in educational theory and research value. We conduct systematic experiments on three representative LLMs, comparing their original versions with ten persona-fine-tuned variants trained on EduPersona. Results show consistent and significant average improvements across all tasks: TASK1 +33.6%, TASK2 +30.6%, and TASK3 +14.9%. These improvements highlight the dataset's effectiveness and research value, while also revealing the heterogeneous difficulty of persona modeling. In summary, EduPersona delivers the first classroom benchmark centered on subjective abilities, establishes a decoupled and verifiable research paradigm, and we will open-source both the dataset and the framework to support the broader research community in advancing trustworthy and human-like AI for education.

---

### On decomposability and subdifferential of the tensor nuclear norm

Authors: Jiewen Guan, Bo Jiang, Zhening Li
Published: 2025-10-06
Categories: math.OC, stat.ML
arXiv: https://arxiv.org/abs/2510.04647v2

We study the decomposability and the subdifferential of the tensor nuclear norm. Both concepts are well understood and widely applied in matrices but remain unclear for higher-order tensors. We show that the tensor nuclear norm admits a full decomposability over specific subspaces and determine the largest possible subspaces that allow the full decomposability. We derive novel inclusions of the subdifferential of the tensor nuclear norm and study its subgradients in a variety of subspaces of interest. All the results hold for tensors of an arbitrary order. As an immediate application, we establish the statistical performance of the tensor robust principal component analysis, the first such result for tensors of an arbitrary order.

---

### Predictive Feature Caching for Training-free Acceleration of Molecular Geometry Generation

Authors: Johanna Sommer, John Rachwan, Nils Fleischmann, Stephan Günnemann, Bertrand Charpentier
Published: 2025-10-06
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.04646v2

Flow matching models generate high-fidelity molecular geometries but incur significant computational costs during inference, requiring hundreds of network evaluations. This inference overhead becomes the primary bottleneck when such models are employed in practice to sample large numbers of molecular candidates. This work discusses a training-free caching strategy that accelerates molecular geometry generation by predicting intermediate hidden states across solver steps. The proposed method operates directly on the SE(3)-equivariant backbone, is compatible with pretrained models, and is orthogonal to existing training-based accelerations and system-level optimizations. Experiments on the GEOM-Drugs dataset demonstrate that caching achieves a twofold reduction in wall-clock inference time at matched sample quality and a speedup of up to 3x compared to the base model with minimal sample quality degradation. Because these gains compound with other optimizations, applying caching alongside other general, lossless optimizations yield as much as a 7x speedup.

---

### Do Superpixel Segmentation Methods Influence Deforestation Image Classification?

Authors: Hugo Resende, Fabio A. Faria, Eduardo B. Neto, Isabela Borlido, Victor Sundermann, Silvio Jamil F. Guimarães, Álvaro L. Fazenda
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04645v1

Image segmentation is a crucial step in various visual applications, including environmental monitoring through remote sensing. In the context of the ForestEyes project, which combines citizen science and machine learning to detect deforestation in tropical forests, image segments are used for labeling by volunteers and subsequent model training. Traditionally, the Simple Linear Iterative Clustering (SLIC) algorithm is adopted as the segmentation method. However, recent studies have indicated that other superpixel-based methods outperform SLIC in remote sensing image segmentation, and might suggest that they are more suitable for the task of detecting deforested areas. In this sense, this study investigated the impact of the four best segmentation methods, together with SLIC, on the training of classifiers for the target application. Initially, the results showed little variation in performance among segmentation methods, even when selecting the top five classifiers using the PyCaret AutoML library. However, by applying a classifier fusion approach (ensemble of classifiers), noticeable improvements in balanced accuracy were observed, highlighting the importance of both the choice of segmentation method and the combination of machine learning-based models for deforestation detection tasks.

---

### QuantAgents: Towards Multi-agent Financial System via Simulated Trading

Authors: Xiangyu Li, Yawen Zeng, Xiaofen Xing, Jin Xu, Xiangmin Xu
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04643v1

In this paper, our objective is to develop a multi-agent financial system that incorporates simulated trading, a technique extensively utilized by financial professionals. While current LLM-based agent models demonstrate competitive performance, they still exhibit significant deviations from real-world fund companies. A critical distinction lies in the agents' reliance on ``post-reflection'', particularly in response to adverse outcomes, but lack a distinctly human capability: long-term prediction of future trends. Therefore, we introduce QuantAgents, a multi-agent system integrating simulated trading, to comprehensively evaluate various investment strategies and market scenarios without assuming actual risks. Specifically, QuantAgents comprises four agents: a simulated trading analyst, a risk control analyst, a market news analyst, and a manager, who collaborate through several meetings. Moreover, our system incentivizes agents to receive feedback on two fronts: performance in real-world markets and predictive accuracy in simulated trading. Extensive experiments demonstrate that our framework excels across all metrics, yielding an overall return of nearly 300% over the three years (https://quantagents.github.io/).

---

### Evaluating LLMs for Demographic-Targeted Social Bias Detection: A Comprehensive Benchmark Study

Authors: Ayan Majumdar, Feihao Chen, Jinghui Li, Xiaozhen Wang
Published: 2025-10-06
Categories: cs.CL, cs.CY, cs.LG
arXiv: https://arxiv.org/abs/2510.04641v3

Large-scale web-scraped text corpora used to train general-purpose AI models often contain harmful demographic-targeted social biases, creating a regulatory need for data auditing and developing scalable bias-detection methods. Although prior work has investigated biases in text datasets and related detection methods, these studies remain narrow in scope. They typically focus on a single content type (e.g., hate speech), cover limited demographic axes, overlook biases affecting multiple demographics simultaneously, and analyze limited techniques. Consequently, practitioners lack a holistic understanding of the strengths and limitations of recent large language models (LLMs) for automated bias detection. In this study, we conduct a comprehensive benchmark study on English texts to assess the ability of LLMs in detecting demographic-targeted social biases. To align with regulatory requirements, we frame bias detection as a multi-label task of detecting targeted identities using a demographic-focused taxonomy. We then systematically evaluate models across scales and techniques, including prompting, in-context learning, and fine-tuning. Using twelve datasets spanning diverse content types and demographics, our study demonstrates the promise of fine-tuned smaller models for scalable detection. However, our analyses also expose persistent gaps across demographic axes and multi-demographic targeted biases, underscoring the need for more effective and scalable detection frameworks.

---

### Social Agent: Mastering Dyadic Nonverbal Behavior Generation via Conversational LLM Agents

Authors: Zeyi Zhang, Yanju Zhou, Heyuan Yao, Tenglong Ao, Xiaohang Zhan, Libin Liu
Published: 2025-10-06
Categories: cs.GR, cs.CV
arXiv: https://arxiv.org/abs/2510.04637v1

We present Social Agent, a novel framework for synthesizing realistic and contextually appropriate co-speech nonverbal behaviors in dyadic conversations. In this framework, we develop an agentic system driven by a Large Language Model (LLM) to direct the conversation flow and determine appropriate interactive behaviors for both participants. Additionally, we propose a novel dual-person gesture generation model based on an auto-regressive diffusion model, which synthesizes coordinated motions from speech signals. The output of the agentic system is translated into high-level guidance for the gesture generator, resulting in realistic movement at both the behavioral and motion levels. Furthermore, the agentic system periodically examines the movements of interlocutors and infers their intentions, forming a continuous feedback loop that enables dynamic and responsive interactions between the two participants. User studies and quantitative evaluations show that our model significantly improves the quality of dyadic interactions, producing natural, synchronized nonverbal behaviors.

---

### Contrastive Learning Using Graph Embeddings for Domain Adaptation of Language Models in the Process Industry

Authors: Anastasia Zhukova, Jonas Lührs, Christian E. Lobmüller, Bela Gipp
Published: 2025-10-06
Categories: cs.CL, cs.IR
arXiv: https://arxiv.org/abs/2510.04631v2

Recent trends in NLP utilize knowledge graphs (KGs) to enhance pretrained language models by incorporating additional knowledge from the graph structures to learn domain-specific terminology or relationships between documents that might otherwise be overlooked. This paper explores how SciNCL, a graph-aware neighborhood contrastive learning methodology originally designed for scientific publications, can be applied to the process industry domain, where text logs contain crucial information about daily operations and are often structured as sparse KGs. Our experiments demonstrate that language models fine-tuned with triplets derived from graph embeddings (GE) outperform a state-of-the-art mE5-large text encoder by 9.8-14.3% (5.45-7.96p) on the proprietary process industry text embedding benchmark (PITEB) while having 3 times fewer parameters.

---

### SFANet: Spatial-Frequency Attention Network for Deepfake Detection

Authors: Vrushank Ahire, Aniruddh Muley, Shivam Zample, Siddharth Verma, Pranav Menon, Surbhi Madan, Abhinav Dhall
Published: 2025-10-06
Categories: cs.CV, cs.AI, cs.MM
arXiv: https://arxiv.org/abs/2510.04630v1

Detecting manipulated media has now become a pressing issue with the recent rise of deepfakes. Most existing approaches fail to generalize across diverse datasets and generation techniques. We thus propose a novel ensemble framework, combining the strengths of transformer-based architectures, such as Swin Transformers and ViTs, and texture-based methods, to achieve better detection accuracy and robustness. Our method introduces innovative data-splitting, sequential training, frequency splitting, patch-based attention, and face segmentation techniques to handle dataset imbalances, enhance high-impact regions (e.g., eyes and mouth), and improve generalization. Our model achieves state-of-the-art performance when tested on the DFWild-Cup dataset, a diverse subset of eight deepfake datasets. The ensemble benefits from the complementarity of these approaches, with transformers excelling in global feature extraction and texturebased methods providing interpretability. This work demonstrates that hybrid models can effectively address the evolving challenges of deepfake detection, offering a robust solution for real-world applications.

---

### A Spatial-Spectral-Frequency Interactive Network for Multimodal Remote Sensing Classification

Authors: Hao Liu, Yunhao Gao, Wei Li, Mingyang Zhang, Maoguo Gong, Lorenzo Bruzzone
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04628v3

Deep learning-based methods have achieved significant success in remote sensing Earth observation data analysis. Numerous feature fusion techniques address multimodal remote sensing image classification by integrating global and local features. However, these techniques often struggle to extract structural and detail features from heterogeneous and redundant multimodal images. With the goal of introducing frequency domain learning to model key and sparse detail features, this paper introduces the spatial-spectral-frequency interaction network (S$^2$Fin), which integrates pairwise fusion modules across the spatial, spectral, and frequency domains. Specifically, we propose a high-frequency sparse enhancement transformer that employs sparse spatial-spectral attention to optimize the parameters of the high-frequency filter. Subsequently, a two-level spatial-frequency fusion strategy is introduced, comprising an adaptive frequency channel module that fuses low-frequency structures with enhanced high-frequency details, and a high-frequency resonance mask that emphasizes sharp edges via phase similarity. In addition, a spatial-spectral attention fusion module further enhances feature extraction at intermediate layers of the network. Experiments on four benchmark multimodal datasets with limited labeled data demonstrate that S$^2$Fin performs superior classification, outperforming state-of-the-art methods. The code is available at https://github.com/HaoLiu-XDU/SSFin.

---

### Compressed Concatenation of Small Embedding Models

Authors: Mohamed Ayoub Ben Ayad, Michael Dinzinger, Kanishka Ghosh Dastidar, Jelena Mitrovic, Michael Granitzer
Published: 2025-10-06
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.04626v1

Embedding models are central to dense retrieval, semantic search, and recommendation systems, but their size often makes them impractical to deploy in resource-constrained environments such as browsers or edge devices. While smaller embedding models offer practical advantages, they typically underperform compared to their larger counterparts. To bridge this gap, we demonstrate that concatenating the raw embedding vectors of multiple small models can outperform a single larger baseline on standard retrieval benchmarks. To overcome the resulting high dimensionality of naive concatenation, we introduce a lightweight unified decoder trained with a Matryoshka Representation Learning (MRL) loss. This decoder maps the high-dimensional joint representation to a low-dimensional space, preserving most of the original performance without fine-tuning the base models. We also show that while concatenating more base models yields diminishing gains, the robustness of the decoder's representation under compression and quantization improves. Our experiments show that, on a subset of MTEB retrieval tasks, our concat-encode-quantize pipeline recovers 89\% of the original performance with a 48x compression factor when the pipeline is applied to a concatenation of four small embedding models.

---

### Fairness in Repeated Matching: A Maximin Perspective

Authors: Eugene Lim, Tzeh Yuan Neoh, Nicholas Teh
Published: 2025-10-06
Categories: cs.GT, cs.AI, cs.LG, cs.MA, econ.TH
arXiv: https://arxiv.org/abs/2510.04624v1

We study a sequential decision-making model where a set of items is repeatedly matched to the same set of agents over multiple rounds. The objective is to determine a sequence of matchings that either maximizes the utility of the least advantaged agent at the end of all rounds (optimal) or at the end of every individual round (anytime optimal). We investigate the computational challenges associated with finding (anytime) optimal outcomes and demonstrate that these problems are generally computationally intractable. However, we provide approximation algorithms, fixed-parameter tractable algorithms, and identify several special cases whereby the problem(s) can be solved efficiently. Along the way, we also establish characterizations of Pareto-optimal/maximum matchings, which may be of independent interest to works in matching theory and house allocation.

---

### MedPAO: A Protocol-Driven Agent for Structuring Medical Reports

Authors: Shrish Shrinath Vaidya, Gowthamaan Palani, Sidharth Ramesh, Velmurugan Balasubramanian, Minmini Selvam, Gokulraja Srinivasaraja, Ganapathy Krishnamurthi
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04623v1

The deployment of Large Language Models (LLMs) for structuring clinical data is critically hindered by their tendency to hallucinate facts and their inability to follow domain-specific rules. To address this, we introduce MedPAO, a novel agentic framework that ensures accuracy and verifiable reasoning by grounding its operation in established clinical protocols such as the ABCDEF protocol for CXR analysis. MedPAO decomposes the report structuring task into a transparent process managed by a Plan-Act-Observe (PAO) loop and specialized tools. This protocol-driven method provides a verifiable alternative to opaque, monolithic models. The efficacy of our approach is demonstrated through rigorous evaluation: MedPAO achieves an F1-score of 0.96 on the critical sub-task of concept categorization. Notably, expert radiologists and clinicians rated the final structured outputs with an average score of 4.52 out of 5, indicating a level of reliability that surpasses baseline approaches relying solely on LLM-based foundation models. The code is available at: https://github.com/MiRL-IITM/medpao-agent

---

### Forecasting-based Biomedical Time-series Data Synthesis for Open Data and Robust AI

Authors: Youngjoon Lee, Seongmin Cho, Yehhyun Jo, Jinu Gong, Hyunjoo Jenny Lee, Joonhyuk Kang
Published: 2025-10-06
Categories: cs.LG, eess.SP
arXiv: https://arxiv.org/abs/2510.04622v2

The limited data availability due to strict privacy regulations and significant resource demands severely constrains biomedical time-series AI development, which creates a critical gap between data requirements and accessibility. Synthetic data generation presents a promising solution by producing artificial datasets that maintain the statistical properties of real biomedical time-series data without compromising patient confidentiality. While GANs, VAEs, and diffusion models capture global data distributions, forecasting models offer inductive biases tailored for sequential dynamics. We propose a framework for synthetic biomedical time-series data generation based on recent forecasting models that accurately replicates complex electrophysiological signals such as EEG and EMG with high fidelity. These synthetic datasets can be freely shared for open AI development and consistently improve downstream model performance. Numerical results on sleep-stage classification show up to a 3.71\% performance gain with augmentation and a 91.00\% synthetic-only accuracy that surpasses the real-data-only baseline.

---

### Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models

Authors: Qizheng Zhang, Changran Hu, Shubhangi Upasani, Boyuan Ma, Fenglu Hong, Vamsidhar Kamanuru, Jay Rainton, Chen Wu, Mengmeng Ji, Hanchen Li, Urmish Thakker, James Zou et al.
Published: 2025-10-06
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2510.04618v3

Large language model (LLM) applications such as agents and domain-specific reasoning increasingly rely on context adaptation: modifying inputs with instructions, strategies, or evidence, rather than weight updates. Prior approaches improve usability but often suffer from brevity bias, which drops domain insights for concise summaries, and from context collapse, where iterative rewriting erodes details over time. We introduce ACE (Agentic Context Engineering), a framework that treats contexts as evolving playbooks that accumulate, refine, and organize strategies through a modular process of generation, reflection, and curation. ACE prevents collapse with structured, incremental updates that preserve detailed knowledge and scale with long-context models. Across agent and domain-specific benchmarks, ACE optimizes contexts both offline (e.g., system prompts) and online (e.g., agent memory), consistently outperforming strong baselines: +10.6% on agents and +8.6% on finance, while significantly reducing adaptation latency and rollout cost. Notably, ACE could adapt effectively without labeled supervision and instead by leveraging natural execution feedback. On the AppWorld leaderboard, ACE matches the top-ranked production-level agent on the overall average and surpasses it on the harder test-challenge split, despite using a smaller open-source model. These results show that comprehensive, evolving contexts enable scalable, efficient, and self-improving LLM systems with low overhead.

---

### AdaR: A Framework for Equipping LLMs with Adaptive Reasoning

Authors: Zhejian Lai, Xiang Geng, Zhijun Wang, Yang Bai, Jiahuan Li, Rongxiang Weng, Jingang Wang, Xuezhi Cao, Xunliang Cai, Shujian Huang
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04617v3

Mathematical reasoning is a primary indicator of large language models (LLMs) intelligence. However, existing LLMs exhibit failures in robustness and generalization. This paper attributes these deficiencies to spurious reasoning, wherein generated reasoning traces are driven by superficial correlations, leading models to blindly reproduce memorized patterns from the training data. To address this challenge, we propose the AdaR framework to equip LLMs with adaptive reasoning, wherein models establish correct correlations between query templates and reasoning processes. AdaR automatically synthesizes logically equivalent queries by varying variable values and trains models using Reinforcement Learning with Verifiable Rewards (RLVR) to penalize spurious logic while encouraging adaptive logic. To ensure data quality, we extract the problem-solving logic from the original query, generate the corresponding answer through code execution, and then apply a sanity check. Experimental results demonstrate that AdaR achieves substantial improvements in mathematical reasoning while maintaining high data efficiency. Furthermore, even advanced LLMs still exhibit deficiencies in robustness and generalization, which our work effectively mitigates. Our project is available at https://github.com/NJUNLP/AdaR.

---

### Design Process of a Self Adaptive Smart Serious Games Ecosystem

Authors: X. Tao, P. Chen, M. Tsami, F. Khayati, M. Eckert
Published: 2025-10-06
Categories: eess.SY, cs.AI
arXiv: https://arxiv.org/abs/2510.04615v1

This paper outlines the design vision and planned evolution of Blexer v3, a modular and AI-driven rehabilitation ecosystem based on serious games. Building on insights from previous versions of the system, we propose a new architecture that aims to integrate multimodal sensing, real-time reasoning, and intelligent control. The envisioned system will include distinct modules for data collection, user state inference, and gameplay adaptation. Key features such as dynamic difficulty adjustment (DDA) and procedural content generation (PCG) are also considered to support personalized interventions. We present the complete conceptual framework of Blexer v3, which defines the modular structure and data flow of the system. This serves as the foundation for the next phase: the development of a functional prototype and its integration into clinical rehabilitation scenarios.

---

### Accountability Capture: How Record-Keeping to Support AI Transparency and Accountability (Re)shapes Algorithmic Oversight

Authors: Shreya Chappidi, Jennifer Cobbe, Chris Norval, Anjali Mazumder, Jatinder Singh
Published: 2025-10-06
Categories: cs.CY, cs.AI
arXiv: https://arxiv.org/abs/2510.04609v1

Accountability regimes typically encourage record-keeping to enable the transparency that supports oversight, investigation, contestation, and redress. However, implementing such record-keeping can introduce considerations, risks, and consequences, which so far remain under-explored. This paper examines how record-keeping practices bring algorithmic systems within accountability regimes, providing a basis to observe and understand their effects. For this, we introduce, describe, and elaborate 'accountability capture' -- the re-configuration of socio-technical processes and the associated downstream effects relating to record-keeping for algorithmic accountability. Surveying 100 practitioners, we evidence and characterise record-keeping issues in practice, identifying their alignment with accountability capture. We further document widespread record-keeping practices, tensions between internal and external accountability requirements, and evidence of employee resistance to practices imposed through accountability capture. We discuss these and other effects for surveillance, privacy, and data protection, highlighting considerations for algorithmic accountability communities. In all, we show that implementing record-keeping to support transparency in algorithmic accountability regimes can itself bring wider implications -- an issue requiring greater attention from practitioners, researchers, and policymakers alike.

---

### From Imperative to Declarative: Towards LLM-friendly OS Interfaces for Boosted Computer-Use Agents

Authors: Yuan Wang, Mingyu Li, Haibo Chen
Published: 2025-10-06
Categories: cs.OS, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.04607v2

Computer-use agents (CUAs) powered by large language models (LLMs) have emerged as a promising approach to automating computer tasks, yet they struggle with the existing human-oriented OS interfaces - graphical user interfaces (GUIs). GUIs force LLMs to decompose high-level goals into lengthy, error-prone sequences of fine-grained actions, resulting in low success rates and an excessive number of LLM calls. We propose Declarative Model Interface (DMI), an abstraction that transforms existing GUIs into three declarative primitives: access, state, and observation, thereby providing novel OS interfaces tailored for LLM agents. Our key idea is policy-mechanism separation: LLMs focus on high-level semantic planning (policy) while DMI handles low-level navigation and interaction (mechanism). DMI does not require modifying the application source code or relying on application programming interfaces (APIs). We evaluate DMI with Microsoft Office Suite (Word, PowerPoint, Excel) on Windows. Integrating DMI into a leading GUI-based agent baseline improves task success rates by 67% and reduces interaction steps by 43.5%. Notably, DMI completes over 61% of successful tasks with a single LLM call.

---

### Closed-Form Last Layer Optimization

Authors: Alexandre Galashov, Nathaël Da Costa, Liyuan Xu, Philipp Hennig, Arthur Gretton
Published: 2025-10-06
Categories: cs.LG, stat.ML
arXiv: https://arxiv.org/abs/2510.04606v2

Neural networks are typically optimized with variants of stochastic gradient descent. Under a squared loss, however, the optimal solution to the linear last layer weights is known in closed-form. We propose to leverage this during optimization, treating the last layer as a function of the backbone parameters, and optimizing solely for these parameters. We show this is equivalent to alternating between gradient descent steps on the backbone and closed-form updates on the last layer. We adapt the method for the setting of stochastic gradient descent, by trading off the loss on the current batch against the accumulated information from previous batches. We provide theoretical analyses showing convergence of the method to an optimal solution in the neural tangent kernel regime, as well as quantifying the gains compared to standard SGD in a one-step analysis. Finally, we demonstrate the effectiveness of our approach compared with SGD and Adam on a squared loss in several regression tasks, including neural operators and causal inference.

---

### Wasserstein Gradient Flows for Scalable and Regularized Barycenter Computation

Authors: Eduardo Fernandes Montesuma, Yassir Bendou, Mike Gartrell
Published: 2025-10-06
Categories: stat.ML, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.04602v4

Wasserstein barycenters provide a principled approach for aggregating probability measures, while preserving the geometry of their ambient space. Existing discrete methods are not because as they assume access to the complete set of samples from the input measures. Meanwhile, neural network approaches do scale well, but rely on complex optimization problems and cannot easily incorporate label information. We address these limitations through gradient flows in the space of probability measures. Through time discretization, we achieve a scalable algorithm that i) relies on mini-batch optimal transport, ii) accepts modular regularization through task-aware functions, and iii) seamlessly integrates supervised information into the ground-cost. We empirically validate our approach on domain adaptation benchmarks that span computer vision, neuroscience, and chemical engineering. Our method establishes a new state-of-the-art Wasserstein barycenter solver, with labeled barycenters consistently outperforming unlabeled ones. Our code at https://github.com/SigmaNova/barycentric-gradient-flows

---

### FedSRD: Sparsify-Reconstruct-Decompose for Communication-Efficient Federated Large Language Models Fine-Tuning

Authors: Guochen Yan, Luyuan Xie, Qingni Shen, Yuejian Fang, Zhonghai Wu
Published: 2025-10-06
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.04601v3

The current paradigm of training large language models (LLMs) on public available Web data is becoming unsustainable as high-quality data sources in specialized domains near exhaustion. Federated Learning (FL) emerges as a practical solution for the next generation of AI on a decentralized Web, enabling privacy-preserving collaborative fine-tuning on decentralized private data. While Low-Rank Adaptation (LoRA) is standard for efficient fine-tuning, its federated application faces a critical bottleneck: communication overhead under heterogeneous network conditions. Structural redundancy in LoRA parameters increases communication costs and causes aggregation conflicts. To address this, we propose FedSRD, a Sparsify-Reconstruct-Decompose framework for communication-efficient federated LLM fine-tuning. We introduce importance-aware sparsification to reduce the upload parameter count while preserving the structural integrity of LoRA updates. The server aggregates updates in full-rank space to mitigate conflicts, then decomposes the global update into a sparse low-rank format for broadcast, ensuring a symmetrically efficient cycle. We also propose an efficient variant, FedSRD-e, to reduce computational overhead. Experiments on 10 benchmarks show our framework significantly reduces communication costs by up to 90\% while improving performance on heterogeneous client data.

---

### SpikingMamba: Towards Energy-Efficient Large Language Models via Knowledge Distillation from Mamba

Authors: Yulong Huang, Jianxiong Tang, Chao Wang, Ziyi Wang, Jianguo Zhang, Zhichao Lu, Bojun Cheng, Luziwei Leng
Published: 2025-10-06
Categories: cs.NE
arXiv: https://arxiv.org/abs/2510.04595v2

Large Language Models (LLMs) have achieved remarkable performance across tasks but remain energy-intensive due to dense matrix operations. Spiking neural networks (SNNs) improve energy efficiency by replacing dense matrix multiplications with sparse accumulations. Their sparse spike activity enables efficient LLMs deployment on edge devices. However, prior SNN-based LLMs often sacrifice performance for efficiency, and recovering accuracy typically requires full pretraining, which is costly and impractical. To address this, we propose SpikingMamba, an energy-efficient SNN-based LLMs distilled from Mamba that improves energy efficiency with minimal accuracy sacrifice. SpikingMamba integrates two key components: (a) SI-LIF, a signed-integer spiking neuron that preserves semantic polarity through signed multi-level spike representations. (b) A training-exclusive Smoothed Gradient Compensation (SGC) path mitigating quantization loss while preserving spike-driven efficiency. We employ a single-stage distillation strategy to transfer the zero-shot ability of pretrained Mamba and further enhance it via reinforcement learning (RL). Experiments show that SpikingMamba-1.3B achieves a 4.76$\times$ energy benefit, with only a 4.78\% zero-shot accuracy gap compared to the original Mamba. The model achieves a further 2.55\% accuracy improvement after RL, narrowing the performance gap from 4.78\% to 2.23\%. Code is available at: https://github.com/HuuYuLong/SpikingMamba .

---

### Data-Driven Adaptive PID Control Based on Physics-Informed Neural Networks

Authors: Junsei Ito, Yasuaki Wasa
Published: 2025-10-06
Categories: eess.SY, cs.LG
arXiv: https://arxiv.org/abs/2510.04591v2

This article proposes a data-driven PID controller design based on the principle of adaptive gain optimization, leveraging Physics-Informed Neural Networks (PINNs) generated for predictive modeling purposes. The proposed control design method utilizes gradients of the PID gain optimization, achieved through the automatic differentiation of PINNs, to apply model predictive control using a cost function based on tracking error and control inputs. By optimizing PINNs-based PID gains, the method achieves adaptive gain tuning that ensures stability while accounting for system nonlinearities. The proposed method features a systematic framework for integrating PINNs-based models of dynamical control systems into closed-loop control systems, enabling direct application to PID control design. A series of numerical experiments is conducted to demonstrate the effectiveness of the proposed method from the control perspectives based on both time and frequency domains.

---

### Perfect AI Mimicry and the Epistemology of Consciousness: A Solipsistic Dilemma

Authors: Shurui Li
Published: 2025-10-06
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.04588v1

Rapid advances in artificial intelligence necessitate a re-examination of the epistemological foundations upon which we attribute consciousness. As AI systems increasingly mimic human behavior and interaction with high fidelity, the concept of a "perfect mimic"-an entity empirically indistinguishable from a human through observation and interaction-shifts from hypothetical to technologically plausible. This paper argues that such developments pose a fundamental challenge to the consistency of our mind-recognition practices. Consciousness attributions rely heavily, if not exclusively, on empirical evidence derived from behavior and interaction. If a perfect mimic provides evidence identical to that of humans, any refusal to grant it equivalent epistemic status must invoke inaccessible factors, such as qualia, substrate requirements, or origin. Selectively invoking such factors risks a debilitating dilemma: either we undermine the rational basis for attributing consciousness to others (epistemological solipsism), or we accept inconsistent reasoning. I contend that epistemic consistency demands we ascribe the same status to empirically indistinguishable entities, regardless of metaphysical assumptions. The perfect mimic thus acts as an epistemic mirror, forcing critical reflection on the assumptions underlying intersubjective recognition in light of advancing AI. This analysis carries significant implications for theories of consciousness and ethical frameworks concerning artificial agents.

---

### Pathology-CoT: Learning Visual Chain-of-Thought Agent from Expert Whole Slide Image Diagnosis Behavior

Authors: Sheng Wang, Ruiming Wu, Charles Herndon, Yihang Liu, Shunsuke Koga, Jeanne Shen, Zhi Huang
Published: 2025-10-06
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.04587v2

Diagnosing a whole-slide image is an interactive, multi-stage process of changing magnification and moving between fields. Although recent pathology foundation models demonstrated superior performances, practical agentic systems that decide what field to examine next, adjust magnification, and deliver explainable diagnoses are still lacking. Such limitation is largely bottlenecked by data: scalable, clinically aligned supervision of expert viewing behavior that is tacit and experience-based, not documented in textbooks or internet, and therefore absent from LLM training. Here we introduce a framework designed to address this challenge through three key breakthroughs. First, the AI Session Recorder seamlessly integrates with standard whole-slide image viewers to unobtrusively record routine navigation and convert the viewer logs into standardized behavioral commands and bounding boxes. Second, a lightweight human-in-the-loop review turns AI-drafted rationales for behavioral commands into the Pathology-CoT dataset, a form of paired "where to look" and "why it matters", enabling six-fold faster labeling compared to manual constructing such Chain-of-Thought dataset. Using this behavioral data, we build Pathology-o3, a two-stage agent that first proposes important ROIs and then performs behavior-guided reasoning. On the gastrointestinal lymph-node metastasis detection task, our method achieved 100 recall on the internal validation from Stanford Medicine and 97.6 recall on an independent external validation from Sweden, exceeding the state-of-the-art OpenAI o3 model and generalizing across backbones. To our knowledge, Pathology-CoT constitutes one of the first behavior-grounded agentic systems in pathology. Turning everyday viewer logs into scalable, expert-validated supervision, our framework makes agentic pathology practical and establishes a path to human-aligned, upgradeable clinical AI.

---

### Robustness assessment of large audio language models in multiple-choice evaluation

Authors: Fernando López, Santosh Kesiraju, Jordi Luque
Published: 2025-10-06
Categories: cs.CL, cs.SD, eess.AS
arXiv: https://arxiv.org/abs/2510.04584v2

Recent advances in large audio language models (LALMs) have primarily been assessed using a multiple-choice question answering (MCQA) framework. However, subtle changes, such as shifting the order of choices, result in substantially different results. Existing MCQA frameworks do not account for this variability and report a single accuracy number per benchmark or category. We dive into the MCQA evaluation framework and conduct a systematic study spanning three benchmarks (MMAU, MMAR and MMSU) and four models: Audio Flamingo 2, Audio Flamingo 3, Qwen2.5-Omni-7B-Instruct, and Kimi-Audio-7B-Instruct. Our findings indicate that models are sensitive not only to the ordering of choices, but also to the paraphrasing of the question and the choices. Finally, we propose a simpler evaluation protocol and metric that account for subtle variations and provide a more detailed evaluation report of LALMs within the MCQA framework.

---