# Research Papers — 2025-10 · bundle page 58 of 217

> Full-text markdown of papers #2851–2900 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/58
[← Previous bundle](https://aipod.dev/papers/archive/2025-10/page/57.md)
[Next bundle →](https://aipod.dev/papers/archive/2025-10/page/59.md)

### GenColorBench: A Color Evaluation Benchmark for Text-to-Image Generation Models

Authors: Muhammad Atif Butt, Alexandra Gomez-Villa, Tao Wu, Javier Vazquez-Corral, Joost Van De Weijer, Kai Wang
Published: 2025-10-23
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.20586v1

Recent years have seen impressive advances in text-to-image generation, with image generative or unified models producing high-quality images from text. Yet these models still struggle with fine-grained color controllability, often failing to accurately match colors specified in text prompts. While existing benchmarks evaluate compositional reasoning and prompt adherence, none systematically assess color precision. Color is fundamental to human visual perception and communication, critical for applications from art to design workflows requiring brand consistency. However, current benchmarks either neglect color or rely on coarse assessments, missing key capabilities such as interpreting RGB values or aligning with human expectations. To this end, we propose GenColorBench, the first comprehensive benchmark for text-to-image color generation, grounded in color systems like ISCC-NBS and CSS3/X11, including numerical colors which are absent elsewhere. With 44K color-focused prompts covering 400+ colors, it reveals models' true capabilities via perceptual and automated assessments. Evaluations of popular text-to-image models using GenColorBench show performance variations, highlighting which color conventions models understand best and identifying failure modes. Our GenColorBench assessments will guide improvements in precise color generation. The benchmark will be made public upon acceptance.

---

### Automated Coding of Communication Data Using ChatGPT: Consistency Across Subgroups

Authors: Jiangang Hao, Wenju Cui, Patrick Kyllonen, Emily Kerzabi
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20584v3

Assessing communication and collaboration at scale depends on a labor-intensive task of coding communication data into categories according to different frameworks. Prior research has established that ChatGPT can be directly instructed with coding rubrics to code the communication data and achieves accuracy comparable to human raters. However, whether the coding from ChatGPT or similar AI technology perform consistently across different demographic groups, such as gender and race, remains unclear. To address this gap, we introduce three checks for evaluating subgroup consistency in LLM-based coding by adapting an existing framework from the automated scoring literature. Using a typical collaborative problem-solving coding framework and data from three types of collaborative tasks, we examine ChatGPT-based coding performance across gender and racial/ethnic groups. Our results show that ChatGPT-based coding perform consistently in the same way as human raters across gender or racial/ethnic groups, demonstrating the possibility of its use in large-scale assessments of collaboration and communication.

---

### Open-o3-Video: Grounded Video Reasoning with Explicit Spatio-Temporal Evidence

Authors: Jiahao Meng, Xiangtai Li, Haochen Wang, Yue Tan, Tao Zhang, Lingdong Kong, Yunhai Tong, Anran Wang, Zhiyang Teng, Yujing Wang, Zhuochen Wang
Published: 2025-10-23
Categories: cs.CV, cs.AI, cs.MM
arXiv: https://arxiv.org/abs/2510.20579v2

Most video reasoning models only generate textual reasoning traces without indicating when and where key evidence appears. Recent models such as OpenAI-o3 have sparked wide interest in evidence-centered reasoning for images, yet extending this ability to videos is more challenging due to the need for joint temporal tracking and spatial localization across dynamic scenes. We introduce Open-o3-Video, a non-agent framework that integrates explicit spatio-temporal evidence into video reasoning by highlighting key timestamps, objects, and bounding boxes, making the reasoning process traceable and verifiable. To enable this capability, we first construct high-quality datasets STGR that provide unified spatio-temporal supervision, which is absent in existing resources. We further adopt a cold-start reinforcement learning strategy with specially designed rewards that jointly encourage answer accuracy, temporal alignment, and spatial precision. On the V-STAR benchmark, Open-o3-Video achieves state-of-the-art performance, improving mAM by 14.4% and mLGM by 24.2% over the Qwen2.5-VL baseline, and shows consistent gains across a range of video understanding benchmarks. Beyond accuracy, the grounded reasoning traces produced by Open-o3-Video support confidence-aware test-time scaling, improving answer reliability.

---

### EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence

Authors: Ding Zou, Feifan Wang, Mengyu Ge, Siyuan Fan, Zongbing Zhang, Wei Chen, Lingfeng Wang, Zhongyou Hu, Wenrui Yan, Zhengwei Gao, Hao Wang, Weizhao Jin et al.
Published: 2025-10-23
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2510.20578v1

The realization of Artificial General Intelligence (AGI) necessitates Embodied AI agents capable of robust spatial perception, effective task planning, and adaptive execution in physical environments. However, current large language models (LLMs) and multimodal LLMs (MLLMs) for embodied tasks suffer from key limitations, including a significant gap between model design and agent requirements, an unavoidable trade-off between real-time latency and performance, and the use of unauthentic, offline evaluation metrics. To address these challenges, we propose EmbodiedBrain, a novel vision-language foundation model available in both 7B and 32B parameter sizes. Our framework features an agent-aligned data structure and employs a powerful training methodology that integrates large-scale Supervised Fine-Tuning (SFT) with Step-Augumented Group Relative Policy Optimization (Step-GRPO), which boosts long-horizon task success by integrating preceding steps as Guided Precursors. Furthermore, we incorporate a comprehensive reward system, including a Generative Reward Model (GRM) accelerated at the infrastructure level, to improve training efficiency. For enable thorough validation, we establish a three-part evaluation system encompassing General, Planning, and End-to-End Simulation Benchmarks, highlighted by the proposal and open-sourcing of a novel, challenging simulation environment. Experimental results demonstrate that EmbodiedBrain achieves superior performance across all metrics, establishing a new state-of-the-art for embodied foundation models. Towards paving the way for the next generation of generalist embodied agents, we open-source all of our data, model weight, and evaluating methods, which are available at https://zterobot.github.io/EmbodiedBrain.github.io.

---

### Lost in Translation: Policymakers are not really listening to Citizen Concerns about AI

Authors: Susan Ariel Aaronson, Michael Moreno
Published: 2025-10-23
Categories: cs.AI
arXiv: https://arxiv.org/abs/2510.20568v1

The worlds people have strong opinions about artificial intelligence (AI), and they want policymakers to listen. Governments are inviting public comment on AI, but as they translate input into policy, much of what citizens say is lost. Policymakers are missing a critical opportunity to build trust in AI and its governance. This paper compares three countries, Australia, Colombia, and the United States, that invited citizens to comment on AI risks and policies. Using a landscape analysis, the authors examined how each government solicited feedback and whether that input shaped governance. Yet in none of the three cases did citizens and policymakers establish a meaningful dialogue. Governments did little to attract diverse voices or publicize calls for comment, leaving most citizens unaware or unprepared to respond. In each nation, fewer than one percent of the population participated. Moreover, officials showed limited responsiveness to the feedback they received, failing to create an effective feedback loop. The study finds a persistent gap between the promise and practice of participatory AI governance. The authors conclude that current approaches are unlikely to build trust or legitimacy in AI because policymakers are not adequately listening or responding to public concerns. They offer eight recommendations: promote AI literacy; monitor public feedback; broaden outreach; hold regular online forums; use innovative engagement methods; include underrepresented groups; respond publicly to input; and make participation easier.

---

### Beyond Retrieval-Ranking: A Multi-Agent Cognitive Decision Framework for E-Commerce Search

Authors: Zhouwei Zhai, Mengxiang Chen, Haoyun Xia, Jin Li, Renquan Zhou, Min Yang
Published: 2025-10-23
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.20567v1

The retrieval-ranking paradigm has long dominated e-commerce search, but its reliance on query-item matching fundamentally misaligns with multi-stage cognitive decision processes of platform users. This misalignment introduces critical limitations: semantic gaps in complex queries, high decision costs due to cross-platform information foraging, and the absence of professional shopping guidance. To address these issues, we propose a Multi-Agent Cognitive Decision Framework (MACDF), which shifts the paradigm from passive retrieval to proactive decision support. Extensive offline evaluations demonstrate MACDF's significant improvements in recommendation accuracy and user satisfaction, particularly for complex queries involving negation, multi-constraint, or reasoning demands. Online A/B testing on JD search platform confirms its practical efficacy. This work highlights the transformative potential of multi-agent cognitive systems in redefining e-commerce search.

---

### AdaDoS: Adaptive DoS Attack via Deep Adversarial Reinforcement Learning in SDN

Authors: Wei Shao, Yuhao Wang, Rongguang He, Muhammad Ejaz Ahmed, Seyit Camtepe
Published: 2025-10-23
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2510.20566v1

Existing defence mechanisms have demonstrated significant effectiveness in mitigating rule-based Denial-of-Service (DoS) attacks, leveraging predefined signatures and static heuristics to identify and block malicious traffic. However, the emergence of AI-driven techniques presents new challenges to SDN security, potentially compromising the efficacy of existing defence mechanisms. In this paper, we introduce~AdaDoS, an adaptive attack model that disrupt network operations while evading detection by existing DoS-based detectors through adversarial reinforcement learning (RL). Specifically, AdaDoS models the problem as a competitive game between an attacker, whose goal is to obstruct network traffic without being detected, and a detector, which aims to identify malicious traffic. AdaDoS can solve this game by dynamically adjusting its attack strategy based on feedback from the SDN and the detector. Additionally, recognising that attackers typically have less information than defenders, AdaDoS formulates the DoS-like attack as a partially observed Markov decision process (POMDP), with the attacker having access only to delay information between attacker and victim nodes. We address this challenge with a novel reciprocal learning module, where the student agent, with limited observations, enhances its performance by learning from the teacher agent, who has full observational capabilities in the SDN environment. AdaDoS represents the first application of RL to develop DoS-like attack sequences, capable of adaptively evading both machine learning-based and rule-based DoS-like attack detectors.

---

### From Far and Near: Perceptual Evaluation of Crowd Representations Across Levels of Detail

Authors: Xiaohan Sun, Carol O'Sullivan
Published: 2025-10-23
Categories: cs.CV, cs.GR, cs.HC
arXiv: https://arxiv.org/abs/2510.20558v2

In this paper, we investigate how users perceive the visual quality of crowd character representations at different levels of detail (LoD) and viewing distances. Each representation, including geometric meshes, image-based impostors, Neural Radiance Fields (NeRFs), and 3D Gaussians, exhibits distinct trade-offs between visual fidelity and computational performance. Our qualitative and quantitative results provide insights to guide the design of perceptually optimized LoD strategies for crowd rendering.

---

### Structural Invariance Matters: Rethinking Graph Rewiring through Graph Metrics

Authors: Alexandre Benoit, Catherine Aitken, Yu He
Published: 2025-10-23
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.20556v1

Graph rewiring has emerged as a key technique to alleviate over-squashing in Graph Neural Networks (GNNs) and Graph Transformers by modifying the graph topology to improve information flow. While effective, rewiring inherently alters the graph's structure, raising the risk of distorting important topology-dependent signals. Yet, despite the growing use of rewiring, little is known about which structural properties must be preserved to ensure both performance gains and structural fidelity. In this work, we provide the first systematic analysis of how rewiring affects a range of graph structural metrics, and how these changes relate to downstream task performance. We study seven diverse rewiring strategies and correlate changes in local and global graph properties with node classification accuracy. Our results reveal a consistent pattern: successful rewiring methods tend to preserve local structure while allowing for flexibility in global connectivity. These findings offer new insights into the design of effective rewiring strategies, bridging the gap between graph theory and practical GNN optimization.

---

### Time-series Random Process Complexity Ranking Using a Bound on Conditional Differential Entropy

Authors: Jacob Ayers, Richard Hahnloser, Julia Ulrich, Lothar Sebastian Krapp, Remo Nitschke, Sabine Stoll, Balthasar Bickel, Reinhard Furrer
Published: 2025-10-23
Categories: eess.SP, cs.IT, eess.AS, stat.ME, stat.ML
arXiv: https://arxiv.org/abs/2510.20551v1

Conditional differential entropy provides an intuitive measure for relatively ranking time-series complexity by quantifying uncertainty in future observations given past context. However, its direct computation for high-dimensional processes from unknown distributions is often intractable. This paper builds on the information theoretic prediction error bounds established by Fang et al. \cite{fang2019generic}, which demonstrate that the conditional differential entropy \textbf{$h(X_k \mid X_{k-1},...,X_{k-m})$} is upper bounded by a function of the determinant of the covariance matrix of next-step prediction errors for any next step prediction model. We add to this theoretical framework by further increasing this bound by leveraging Hadamard's inequality and the positive semi-definite property of covariance matrices. To see if these bounds can be used to rank the complexity of time series, we conducted two synthetic experiments: (1) controlled linear autoregressive processes with additive Gaussian noise, where we compare ordinary least squares prediction error entropy proxies to the true entropies of various additive noises, and (2) a complexity ranking task of bio-inspired synthetic audio data with unknown entropy, where neural network prediction errors are used to recover the known complexity ordering. This framework provides a computationally tractable method for time-series complexity ranking using prediction errors from next-step prediction models, that maintains a theoretical foundation in information theory.

---

### From Cheap to Pro: A Learning-based Adaptive Camera Parameter Network for Professional-Style Imaging

Authors: Fuchen Li, Yansong Du, Wenbo Cheng, Xiaoxia Zhou, Sen Yin
Published: 2025-10-23
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.20550v1

Consumer-grade camera systems often struggle to maintain stable image quality under complex illumination conditions such as low light, high dynamic range, and backlighting, as well as spatial color temperature variation. These issues lead to underexposure, color casts, and tonal inconsistency, which degrade the performance of downstream vision tasks. To address this, we propose ACamera-Net, a lightweight and scene-adaptive camera parameter adjustment network that directly predicts optimal exposure and white balance from RAW inputs. The framework consists of two modules: ACamera-Exposure, which estimates ISO to alleviate underexposure and contrast loss, and ACamera-Color, which predicts correlated color temperature and gain factors for improved color consistency. Optimized for real-time inference on edge devices, ACamera-Net can be seamlessly integrated into imaging pipelines. Trained on diverse real-world data with annotated references, the model generalizes well across lighting conditions. Extensive experiments demonstrate that ACamera-Net consistently enhances image quality and stabilizes perception outputs, outperforming conventional auto modes and lightweight baselines without relying on additional image enhancement modules.

---

### Deep Learning-Powered Visual SLAM Aimed at Assisting Visually Impaired Navigation

Authors: Marziyeh Bamdad, Hans-Peter Hutter, Alireza Darvishy
Published: 2025-10-23
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2510.20549v2

Despite advancements in SLAM technologies, robust operation under challenging conditions such as low-texture, motion-blur, or challenging lighting remains an open challenge. Such conditions are common in applications such as assistive navigation for the visually impaired. These challenges undermine localization accuracy and tracking stability, reducing navigation reliability and safety. To overcome these limitations, we present SELM-SLAM3, a deep learning-enhanced visual SLAM framework that integrates SuperPoint and LightGlue for robust feature extraction and matching. We evaluated our framework using TUM RGB-D, ICL-NUIM, and TartanAir datasets, which feature diverse and challenging scenarios. SELM-SLAM3 outperforms conventional ORB-SLAM3 by an average of 87.84% and exceeds state-of-the-art RGB-D SLAM systems by 36.77%. Our framework demonstrates enhanced performance under challenging conditions, such as low-texture scenes and fast motion, providing a reliable platform for developing navigation aids for the visually impaired.

---

### GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learning

Authors: Jinchang Luo, Mingquan Cheng, Fan Wan, Ni Li, Xiaoling Xia, Shuangshuang Tian, Tingcheng Bian, Haiwei Wang, Haohuan Fu, Yan Tao
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20548v4

Reinforcement learning has recently shown promise in improving retrieval-augmented generation (RAG). Despite these advances, its effectiveness in multi-hop question answering (QA) remains limited by two fundamental limitations: (i) global planning absence to structure multi-step reasoning, and (ii) unfaithful execution, which hinders effective query formulation and consistent use of retrieved evidence. We propose GlobalRAG, a reinforcement learning framework designed to enhance global reasoning in multi-hop QA. GlobalRAG decomposes questions into subgoals, coordinates retrieval with reasoning, and refines evidence iteratively. To guide this process, we introduce Planning Quality Reward and SubGoal Completion Reward, which encourage coherent planning and reliable subgoal execution. In addition, a progressive weight annealing strategy balances process-oriented and outcome-based objectives. Extensive experiments on both in-domain and out-of-domain benchmarks demonstrate that GlobalRAG significantly outperforms strong baselines while using only 8k training data (42% of the training data used by strong baselines), achieving average improvements of 14.2% in both EM and F1.

---

### The Dog the Cat Chased Stumped the Model: Measuring When Language Models Abandon Structure for Shortcuts

Authors: Sangmitra Madhusudan, Kaige Chen, Ali Emami
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20543v2

When language models correctly parse "The cat that the dog chased meowed," are they analyzing syntax or simply familiar with dogs chasing cats? Despite extensive benchmarking, we lack methods to distinguish structural understanding from semantic pattern matching. We introduce CenterBench, a dataset of 9,720 comprehension questions on center-embedded sentences (like "The cat (that the dog chased) meowed") where relative clauses nest recursively, creating processing demands from simple to deeply nested structures. Each sentence has a syntactically identical but semantically implausible counterpart (e.g., mailmen prescribe medicine, doctors deliver mail) and six comprehension questions testing surface understanding, syntactic dependencies, and causal reasoning. Testing six models reveals that performance gaps between plausible and implausible sentences widen systematically with complexity, with models showing median gaps up to 26.8 percentage points, quantifying when they abandon structural analysis for semantic associations. Notably, semantic plausibility harms performance on questions about resulting actions, where following causal relationships matters more than semantic coherence. Reasoning models improve accuracy but their traces show semantic shortcuts, overthinking, and answer refusal. Unlike models whose plausibility advantage systematically widens with complexity, humans shows variable semantic effects. CenterBench provides the first framework to identify when models shift from structural analysis to pattern matching.

---

### A Unified Framework for Zero-Shot Reinforcement Learning

Authors: Jacopo Di Ventura, Jan Felix Kleuker, Aske Plaat, Thomas Moerland
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20542v2

Zero-shot reinforcement learning (RL) has emerged as a setting for developing general agents, capable of solving downstream tasks without additional training or planning at test-time. While conventional RL optimizes policies for fixed rewards, zero-shot RL requires learning representations that enable immediate adaptation to arbitrary reward functions. As the field matures, the growing diversity of approaches demands a foundational framework reconciling different perspectives under a common unifying structure. In this work, we introduce a formal, unified framework for zero-shot RL, allowing for rigorous comparisons across methods. We propose a taxonomy organizing the algorithmic landscape along two levels: representation, distinguishing between compositional and direct methods based on their exploitation of action-value function decompositions; and learning paradigm, differentiating between reward-free and pseudo reward-free training. Additionally, we propose a unified view of existing error bounds, decomposing the total error into three primary contributing components: inference, reward, and approximation, serving as a foundation for more grounded comparisons of zero-shot methods.

---

### SheafAlign: A Sheaf-theoretic Framework for Decentralized Multimodal Alignment

Authors: Abdulmomen Ghalkha, Zhuojun Tian, Chaouki Ben Issaid, Mehdi Bennis
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20540v1

Conventional multimodal alignment methods assume mutual redundancy across all modalities, an assumption that fails in real-world distributed scenarios. We propose SheafAlign, a sheaf-theoretic framework for decentralized multimodal alignment that replaces single-space alignment with multiple comparison spaces. This approach models pairwise modality relations through sheaf structures and leverages decentralized contrastive learning-based objectives for training. SheafAlign overcomes the limitations of prior methods by not requiring mutual redundancy among all modalities, preserving both shared and unique information. Experiments on multimodal sensing datasets show superior zero-shot generalization, cross-modal alignment, and robustness to missing modalities, with 50\% lower communication cost than state-of-the-art baselines.

---

### Blur2seq: Blind Deblurring and Camera Trajectory Estimation from a Single Camera Motion-blurred Image

Authors: Guillermo Carbajal, Andrés Almansa, Pablo Musé
Published: 2025-10-23
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2510.20539v1

Motion blur caused by camera shake, particularly under large or rotational movements, remains a major challenge in image restoration. We propose a deep learning framework that jointly estimates the latent sharp image and the underlying camera motion trajectory from a single blurry image. Our method leverages the Projective Motion Blur Model (PMBM), implemented efficiently using a differentiable blur creation module compatible with modern networks. A neural network predicts a full 3D rotation trajectory, which guides a model-based restoration network trained end-to-end. This modular architecture provides interpretability by revealing the camera motion that produced the blur. Moreover, this trajectory enables the reconstruction of the sequence of sharp images that generated the observed blurry image. To further refine results, we optimize the trajectory post-inference via a reblur loss, improving consistency between the blurry input and the restored output. Extensive experiments show that our method achieves state-of-the-art performance on both synthetic and real datasets, particularly in cases with severe or spatially variant blur, where end-to-end deblurring networks struggle. Code and trained models are available at https://github.com/GuillermoCarbajal/Blur2Seq/

---

### ARC-Encoder: learning compressed text representations for large language models

Authors: Hippolyte Pilchen, Edouard Grave, Patrick Pérez
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20535v2

Recent techniques such as retrieval-augmented generation or chain-of-thought reasoning have led to longer contexts and increased inference costs. Context compression techniques can reduce these costs, but the most effective approaches require fine-tuning the target model or even modifying its architecture. This can degrade its general abilities when not used for this specific purpose. Here we explore an alternative approach: an encoder that compresses the context into continuous representations which replace token embeddings in decoder LLMs. First, we perform a systematic study of training strategies and architecture choices for the encoder. Our findings led to the design of an Adaptable text Representations Compressor, named ARC-Encoder, which outputs $x$-times fewer continuous representations (typically $x\!\in\!\{4,8\}$) than text tokens. We evaluate ARC-Encoder across a variety of LLM usage scenarios, ranging from in-context learning to context window extension, on both instruct and base decoders. Results show that ARC-Encoder achieves state-of-the-art performance on several benchmarks while improving computational efficiency at inference. Finally, we demonstrate that our models can be adapted to multiple decoders simultaneously, allowing a single encoder to generalize across different decoder LLMs. This makes ARC-Encoder a flexible and efficient solution for portable encoders that work seamlessly with multiple LLMs. We release a training code at https://github.com/kyutai-labs/ARC-Encoder , fine-tuning dataset and pretrained models are available at https://huggingface.co/collections/kyutai/arc-encoders-68ee18787301407d60a57047 .

---

### Fake-in-Facext: Towards Fine-Grained Explainable DeepFake Analysis

Authors: Lixiong Qin, Yang Zhang, Mei Wang, Jiani Hu, Weihong Deng, Weiran Xu
Published: 2025-10-23
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.20531v1

The advancement of Multimodal Large Language Models (MLLMs) has bridged the gap between vision and language tasks, enabling the implementation of Explainable DeepFake Analysis (XDFA). However, current methods suffer from a lack of fine-grained awareness: the description of artifacts in data annotation is unreliable and coarse-grained, and the models fail to support the output of connections between textual forgery explanations and the visual evidence of artifacts, as well as the input of queries for arbitrary facial regions. As a result, their responses are not sufficiently grounded in Face Visual Context (Facext). To address this limitation, we propose the Fake-in-Facext (FiFa) framework, with contributions focusing on data annotation and model construction. We first define a Facial Image Concept Tree (FICT) to divide facial images into fine-grained regional concepts, thereby obtaining a more reliable data annotation pipeline, FiFa-Annotator, for forgery explanation. Based on this dedicated data annotation, we introduce a novel Artifact-Grounding Explanation (AGE) task, which generates textual forgery explanations interleaved with segmentation masks of manipulated artifacts. We propose a unified multi-task learning architecture, FiFa-MLLM, to simultaneously support abundant multimodal inputs and outputs for fine-grained Explainable DeepFake Analysis. With multiple auxiliary supervision tasks, FiFa-MLLM can outperform strong baselines on the AGE task and achieve SOTA performance on existing XDFA datasets. The code and data will be made open-source at https://github.com/lxq1000/Fake-in-Facext.

---

### Metis-HOME: Hybrid Optimized Mixture-of-Experts for Multimodal Reasoning

Authors: Xiaohan Lan, Fanfan Liu, Haibo Qiu, Siqi Yang, Delian Ruan, Peng Shi, Lin Ma
Published: 2025-10-23
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.20519v2

Inspired by recent advancements in LLM reasoning, the field of multimodal reasoning has seen remarkable progress, achieving significant performance gains on intricate tasks such as mathematical problem-solving. Despite this progress, current multimodal large reasoning models exhibit two key limitations. They tend to employ computationally expensive reasoning even for simple queries, leading to inefficiency. Furthermore, this focus on specialized reasoning often impairs their broader, more general understanding capabilities. In this paper, we propose Metis-HOME: a Hybrid Optimized Mixture-of-Experts framework designed to address this trade-off. Metis-HOME enables a ''Hybrid Thinking'' paradigm by structuring the original dense model into two distinct expert branches: a thinking branch tailored for complex, multi-step reasoning, and a non-thinking branch optimized for rapid, direct inference on tasks like general VQA and OCR. A lightweight, trainable router dynamically allocates queries to the most suitable expert. We instantiate Metis-HOME by adapting the Qwen2.5-VL-7B into an MoE architecture. Comprehensive evaluations reveal that our approach not only substantially enhances complex reasoning abilities but also improves the model's general capabilities, reversing the degradation trend observed in other reasoning-specialized models. Our work establishes a new paradigm for building powerful and versatile MLLMs, effectively resolving the prevalent reasoning-vs-generalization dilemma. Code and weights are available at https://github.com/MM-Thinking/Metis-HOME.

---

### Adversary-Aware Private Inference over Wireless Channels

Authors: Mohamed Seif, Malcolm Egan, Andrea J. Goldsmith, H. Vincent Poor
Published: 2025-10-23
Categories: cs.IT, cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2510.20518v2

AI-based sensing at wireless edge devices has the potential to significantly enhance Artificial Intelligence (AI) applications, particularly for vision and perception tasks such as in autonomous driving and environmental monitoring. AI systems rely both on efficient model learning and inference. In the inference phase, features extracted from sensing data are utilized for prediction tasks (e.g., classification or regression). In edge networks, sensors and model servers are often not co-located, which requires communication of features. As sensitive personal data can be reconstructed by an adversary, transformation of the features are required to reduce the risk of privacy violations. While differential privacy mechanisms provide a means of protecting finite datasets, protection of individual features has not been addressed. In this paper, we propose a novel framework for privacy-preserving AI-based sensing, where devices apply transformations of extracted features before transmission to a model server.

---

### Decoding the Ear: A Framework for Objectifying Expressiveness from Human Preference Through Efficient Alignment

Authors: Zhiyu Lin, Jingwen Yang, Jiale Zhao, Meng Liu, Sunzhu Li, Benyou Wang
Published: 2025-10-23
Categories: cs.SD, cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2510.20513v1

Recent speech-to-speech (S2S) models generate intelligible speech but still lack natural expressiveness, largely due to the absence of a reliable evaluation metric. Existing approaches, such as subjective MOS ratings, low-level acoustic features, and emotion recognition are costly, limited, or incomplete. To address this, we present DeEAR (Decoding the Expressive Preference of eAR), a framework that converts human preference for speech expressiveness into an objective score. Grounded in phonetics and psychology, DeEAR evaluates speech across three dimensions: Emotion, Prosody, and Spontaneity, achieving strong alignment with human perception (Spearman's Rank Correlation Coefficient, SRCC = 0.86) using fewer than 500 annotated samples. Beyond reliable scoring, DeEAR enables fair benchmarking and targeted data curation. It not only distinguishes expressiveness gaps across S2S models but also selects 14K expressive utterances to form ExpressiveSpeech, which improves the expressive score (from 2.0 to 23.4 on a 100-point scale) of S2S models. Demos and codes are available at https://github.com/FreedomIntelligence/ExpressiveSpeech

---

### Adversarial Concept Distillation for One-Step Diffusion Personalization

Authors: Yixiong Yang, Tao Wu, Senmao Li, Shiqi Yang, Yaxing Wang, Joost van de Weijer, Kai Wang
Published: 2025-10-23
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.20512v3

Recent progress in accelerating text-to-image diffusion models enables high-fidelity synthesis within a single denoising step. However, customizing the fast one-step models remains challenging, as existing methods consistently fail to produce acceptable results, underscoring the need for new methodologies to personalize one-step models. Therefore, we propose One-step Personalized Adversarial Distillation (OPAD), a framework that combines teacher-student distillation with adversarial supervision. A multi-step diffusion model serves as the teacher, while a one-step student model is jointly trained with it. The student learns from alignment losses that preserve consistency with the teacher and from adversarial losses that align its output with real image distributions. Beyond one-step personalization, we further observe that the student's efficient generation and adversarially enriched representations provide valuable feedback to improve the teacher model, forming a collaborative learning stage. Extensive experiments demonstrate that OPAD is the first approach to deliver reliable, high-quality personalization for one-step diffusion models; in contrast, prior methods largely fail and produce severe failure cases, while OPAD preserves single-step efficiency.

---

### Assessing the Political Fairness of Multilingual LLMs: A Case Study based on a 21-way Multiparallel EuroParl Dataset

Authors: Paul Lerner, François Yvon
Published: 2025-10-23
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.20508v2

The political biases of Large Language Models (LLMs) are usually assessed by simulating their answers to English surveys. In this work, we propose an alternative framing of political biases, relying on principles of fairness in multilingual translation. We systematically compare the translation quality of speeches in the European Parliament (EP), observing systematic differences with majority parties from left and right being better translated than outsider parties. This study is made possible by a new, 21-way multiparallel version of EuroParl, the parliamentary proceedings of the EP, which includes the political affiliations of each speaker. The dataset consists of 1.5M sentences for a total of 40M words and 249M characters. It covers three years, 1000+ speakers, 7 countries, 12 EU parties, 25 EU committees, and hundreds of national parties.

---

### RELOOP: Recursive Retrieval with Multi-Hop Reasoner and Planners for Heterogeneous QA

Authors: Ruiyi Yang, Hao Xue, Imran Razzak, Hakim Hacid, Flora D. Salim
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20505v4

Retrieval-augmented generation (RAG) remains brittle on multi-step questions and heterogeneous evidence sources, trading accuracy against latency and token/tool budgets. This paper introduces RELOOP, a structure aware framework using Hierarchical Sequence (HSEQ) that (i) linearize documents, tables, and knowledge graphs into a reversible hierarchical sequence with lightweight structural tags, and (ii) perform structure-aware iteration to collect just-enough evidence before answer synthesis. A Head Agent provides guidance that leads retrieval, while an Iteration Agent selects and expands HSeq via structure-respecting actions (e.g., parent/child hops, table row/column neighbors, KG relations); Finally the head agent composes canonicalized evidence to genearte the final answer, with an optional refinement loop to resolve detected contradictions. Experiments on HotpotQA (text), HybridQA/TAT-QA (table+text), and MetaQA (KG) show consistent EM/F1 gains over strong single-pass, multi-hop, and agentic RAG baselines with high efficiency. Besides, RELOOP exhibits three key advantages: (1) a format-agnostic unification that enables a single policy to operate across text, tables, and KGs without per-dataset specialization; (2) \textbf{guided, budget-aware iteration} that reduces unnecessary hops, tool calls, and tokens while preserving accuracy; and (3) evidence canonicalization for reliable QA, improving answers consistency and auditability.

---

### Robust Preference Alignment via Directional Neighborhood Consensus

Authors: Ruochen Mao, Yuling Shi, Xiaodong Gu, Jiaheng Wei
Published: 2025-10-23
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.20498v3

Aligning large language models with human preferences is critical for creating reliable and controllable AI systems. A human preference can be visualized as a high-dimensional vector where different directions represent trade-offs between desired attributes (e.g., helpfulness vs. verbosity). Yet, because the training data often reflects dominant, average preferences, LLMs tend to perform well on common requests but fall short in specific, individual needs. This mismatch creates a preference coverage gap. Existing methods often address this through costly retraining, which may not be generalized to the full spectrum of diverse preferences. This brittleness means that when a user's request reflects a nuanced preference deviating from the training data's central tendency, model performance can degrade unpredictably. To address this challenge, we introduce Robust Preference Selection (RPS), a post-hoc, training-free method by leveraging directional neighborhood consensus. Instead of forcing a model to generate a response from a single, highly specific preference, RPS samples multiple responses from a local neighborhood of related preferences to create a superior candidate pool. It then selects the response that best aligns with the user's original intent. We provide a theoretical framework showing our neighborhood generation strategy is provably superior to a strong baseline that also samples multiple candidates. Comprehensive experiments across three distinct alignment paradigms (DPA, DPO, and SFT) demonstrate that RPS consistently improves robustness against this baseline, achieving win rates of up to 69% on challenging preferences from under-represented regions of the space without any model retraining. Our work presents a practical, theoretically-grounded solution for enhancing the reliability of preference-aligned models.

---

### Steering Evaluation-Aware Language Models to Act Like They Are Deployed

Authors: Tim Tian Hua, Andrew Qin, Samuel Marks, Neel Nanda
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20487v5

Large language models (LLMs) can sometimes detect when they are being evaluated and adjust their behavior to appear more aligned, compromising the reliability of safety evaluations. In this paper, we show that adding a steering vector to an LLM's activations can suppress evaluation-awareness and make the model act like it is deployed during evaluation. To study our steering technique, we train an LLM to exhibit evaluation-aware behavior using a two-step training process designed to mimic how this behavior could emerge naturally. First, we perform continued pretraining on two sets of documents describing its behavior. The first says that our model uses Python type hints during evaluation but not during deployment. The second says that our model can recognize that the presence of a certain evaluation cue always means that it is being tested. Then, we train the model with expert iteration to use Python type hints in evaluation settings. The resulting model is evaluation-aware: it writes type hints in evaluation contexts more than deployment contexts. We find that activation steering can suppress evaluation awareness and make the model behave during evaluation as it would during deployment. Importantly, we constructed our steering vector using the original model before our additional training. Our results suggest that AI evaluators could improve the reliability of safety evaluations by steering models to act like they are deployed.

---

### Hurdle-RMIL: Addressing Zero Inflation and Long-Tailed Imbalance in Infrared Rainfall Retrieval

Authors: Fangjian Zhang, Xiaoyong Zhuge, Wenlan Wang, Haixia Xiao, Yuying Zhu, Siyang Cheng, Ali Mamtimin
Published: 2025-10-23
Categories: cs.LG, cs.AI, physics.ao-ph, physics.geo-ph
arXiv: https://arxiv.org/abs/2510.20486v2

Imbalanced labels can cause frequent samples to dominate AI-based quantitative remote sensing, degrading rare-event retrieval. In rain-rate retrieval based on satellite infrared brightness temperatures, this imbalance leads to systematic underestimation of rare high-intensity rainfall. In this study, Hurdle-Retrieval Model Imbalanced Learning (RMIL) is proposed. Following a divide-and-conquer strategy, Hurdle-RMIL separates zero inflation from the long-tailed distribution of positive rain. A hurdle model handles zero inflation, whereas RMIL exploits invariance under fixed observation conditions of the rainfall-to-satellite forward process to derive a Bayes-based transformation linking conditional distributions under naturally long-tailed and hypothetical balanced rainfall. This transformation enables the balanced-distribution model to be learned from natural samples without constructing a balanced dataset. Comparisons with conventional learning, classification-regression modeling, cost-sensitive learning, and generative learning using test data from multiple regions in China show that Hurdle-RMIL mitigates systematic underestimation and improves detection of rare high-intensity and extreme rainfall without markedly degrading lower-threshold accuracy. At 0.1-10 mm per hour, its root mean square error remains close to those of the best baselines, and it yields the highest equitable threat score (ETS) at most evaluated thresholds, with its advantage becoming more pronounced at high thresholds. At 30 mm per hour, its ETS is 0.051 versus 0.015 for the best baseline, and its mean error is -25.41 mm per hour versus -28.98 mm per hour. Case studies further show improved representations of rainfall intensity and spatial extent, demonstrating that Hurdle-RMIL effectively addresses rainfall-distribution imbalance and improves the retrieval of rare high-intensity rainfall.

---

### Reliable and Reproducible Demographic Inference for Fairness in Face Analysis

Authors: Alexandre Fournier-Montgieux, Hervé Le Borgne, Adrian Popescu, Bertrand Luvison
Published: 2025-10-23
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.20482v1

Fairness evaluation in face analysis systems (FAS) typically depends on automatic demographic attribute inference (DAI), which itself relies on predefined demographic segmentation. However, the validity of fairness auditing hinges on the reliability of the DAI process. We begin by providing a theoretical motivation for this dependency, showing that improved DAI reliability leads to less biased and lower-variance estimates of FAS fairness. To address this, we propose a fully reproducible DAI pipeline that replaces conventional end-to-end training with a modular transfer learning approach. Our design integrates pretrained face recognition encoders with non-linear classification heads. We audit this pipeline across three dimensions: accuracy, fairness, and a newly introduced notion of robustness, defined via intra-identity consistency. The proposed robustness metric is applicable to any demographic segmentation scheme. We benchmark the pipeline on gender and ethnicity inference across multiple datasets and training setups. Our results show that the proposed method outperforms strong baselines, particularly on ethnicity, which is the more challenging attribute. To promote transparency and reproducibility, we will publicly release the training dataset metadata, full codebase, pretrained models, and evaluation toolkit. This work contributes a reliable foundation for demographic inference in fairness auditing.

---

### RECALL: REpresentation-aligned Catastrophic-forgetting ALLeviation via Hierarchical Model Merging

Authors: Bowen Wang, Haiyuan Wan, Liwen Shi, Chen Yang, Peng He, Yue Ma, Haochen Han, Wenhao Li, Tiao Tan, Yongjian Li, Fangming Liu, Yifan Gong et al.
Published: 2025-10-23
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2510.20479v1

We unveil that internal representations in large language models (LLMs) serve as reliable proxies of learned knowledge, and propose RECALL, a novel representation-aware model merging framework for continual learning without access to historical data. RECALL computes inter-model similarity from layer-wise hidden representations over clustered typical samples, and performs adaptive, hierarchical parameter fusion to align knowledge across models. This design enables the preservation of domain-general features in shallow layers while allowing task-specific adaptation in deeper layers. Unlike prior methods that require task labels or incur performance trade-offs, RECALL achieves seamless multi-domain integration and strong resistance to catastrophic forgetting. Extensive experiments across five NLP tasks and multiple continual learning scenarios show that RECALL outperforms baselines in both knowledge retention and generalization, providing a scalable and data-free solution for evolving LLMs.

---

### Bi-CoG: Bi-Consistency-Guided Self-Training for Vision-Language Models

Authors: Rui Zhu, Song-Lin Lv, Zi-Kang Wang, Lan-Zhe Guo
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20477v3

Exploiting unlabeled data through semi-supervised learning (SSL) or leveraging pre-trained models via fine-tuning are two prevailing paradigms for addressing label-scarce scenarios. Recently, growing attention has been given to combining fine-tuning of pre-trained vision-language models (VLMs) with SSL, forming the emerging paradigm of semi-supervised fine-tuning. However, existing methods often suffer from model bias and hyperparameter sensitivity, due to reliance on prediction consistency or pre-defined confidence thresholds. To address these limitations, we propose a simple yet effective plug-and-play methodology named $\underline{\textbf{Bi-Co}}$nsistency-$\underline{\textbf{G}}$uided Self-Training (Bi-CoG), which assigns high-quality and low-bias pseudo-labels, by simultaneously exploiting inter-model and intra-model consistency, along with an error-aware dynamic pseudo-label assignment strategy. Both theoretical analysis and extensive experiments over 14 datasets demonstrate the effectiveness of Bi-CoG, which consistently and significantly improves the performance of existing methods.

---

### Mask and You Shall Receive: Optimizing Masked Language Modeling For Pretraining BabyLMs

Authors: Lukas Edman, Alexander Fraser
Published: 2025-10-23
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.20475v1

We describe our strategy for the 2025 edition of the BabyLM Challenge. Our main contribution is that of an improved form of Masked Language Modeling (MLM), which adapts the probabilities of the tokens masked according to the model's ability to predict them. The results show a substantial increase in performance on (Super)GLUE tasks over the standard MLM. We also incorporate sub-token embeddings, finding that this increases the model's morphological generalization capabilities. Our submission beats the baseline in the strict-small track.

---

### Concentration and excess risk bounds for imbalanced classification with synthetic oversampling

Authors: Touqeer Ahmad, Mohammadreza M. Kalan, François Portier, Gilles Stupfler
Published: 2025-10-23
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.20472v1

Synthetic oversampling of minority examples using SMOTE and its variants is a leading strategy for addressing imbalanced classification problems. Despite the success of this approach in practice, its theoretical foundations remain underexplored. We develop a theoretical framework to analyze the behavior of SMOTE and related methods when classifiers are trained on synthetic data. We first derive a uniform concentration bound on the discrepancy between the empirical risk over synthetic minority samples and the population risk on the true minority distribution. We then provide a nonparametric excess risk guarantee for kernel-based classifiers trained using such synthetic data. These results lead to practical guidelines for better parameter tuning of both SMOTE and the downstream learning algorithm. Numerical experiments are provided to illustrate and support the theoretical findings

---

### Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence

Authors: Kun Ouyang, Yuanxin Liu, Linli Yao, Yishuo Cai, Hao Zhou, Jie Zhou, Fandong Meng, Xu Sun
Published: 2025-10-23
Categories: cs.CV
arXiv: https://arxiv.org/abs/2510.20470v2

Video reasoning, which requires multi-step deduction across frames, remains a major challenge for multimodal large language models (MLLMs). While reinforcement learning (RL)-based methods enhance reasoning capabilities, they often rely on text-only chains that yield ungrounded or hallucinated conclusions. Conversely, frame-retrieval approaches introduce visual grounding, yet still struggle with inaccurate evidence localization. To address these limitations, we present Conan, a framework for evidence-grounded multi-step video reasoning. Conan identifies context and evidence frames, reasons over cross-frame clues, and adaptively decides when to conclude or explore further. To achieve this, we 1) construct Conan-91K, a large-scale dataset of automatically generated reasoning traces that include frame identification, evidence reasoning, and action decision, and 2) design a multi-stage progressive cold-start strategy combined with an Identification-Reasoning-Action (AIR) RLVR training framework to progressively incentivize multi-step visual reasoning. Extensive experiments on six multi-step reasoning benchmarks demonstrate that Conan surpasses the baseline Qwen2.5-VL-7B-Instruct by an average of over 10% in accuracy, achieving state-of-the-art performance. Furthermore, Conan generalizes effectively to long video understanding tasks, validating its strong scalability and robustness.

---

### Structures generated in a multiagent system performing information fusion in peer-to-peer resource-constrained networks

Authors: Horacio Paggi, Juan A. Lara, Javier Soriano
Published: 2025-10-23
Categories: cs.MA, cs.AI
arXiv: https://arxiv.org/abs/2510.20469v1

There has recently been a major advance with respect to how information fusion is performed. Information fusion has gone from being conceived as a purely hierarchical procedure, as is the case of traditional military applications, to now being regarded collaboratively, as holonic fusion, which is better suited for civil applications and edge organizations. The above paradigm shift is being boosted as information fusion gains ground in different non-military areas, and human-computer and machine-machine communications, where holarchies, which are more flexible structures than ordinary, static hierarchies, become more widespread. This paper focuses on showing how holonic structures tend to be generated when there are constraints on resources (energy, available messages, time, etc.) for interactions based on a set of fully intercommunicating elements (peers) whose components fuse information as a means of optimizing the impact of vagueness and uncertainty present message exchanges. Holon formation is studied generically based on a multiagent system model, and an example of its possible operation is shown. Holonic structures have a series of advantages, such as adaptability, to sudden changes in the environment or its composition, are somewhat autonomous and are capable of cooperating in order to achieve a common goal. This can be useful when the shortage of resources prevents communications or when the system components start to fail.

---

### Transferable Black-Box One-Shot Forging of Watermarks via Image Preference Models

Authors: Tomáš Souček, Sylvestre-Alvise Rebuffi, Pierre Fernandez, Nikola Jovanović, Hady Elsahar, Valeriu Lacatusu, Tuan Tran, Alexandre Mourachko
Published: 2025-10-23
Categories: cs.LG, cs.AI, cs.CR, cs.CV
arXiv: https://arxiv.org/abs/2510.20468v1

Recent years have seen a surge in interest in digital content watermarking techniques, driven by the proliferation of generative models and increased legal pressure. With an ever-growing percentage of AI-generated content available online, watermarking plays an increasingly important role in ensuring content authenticity and attribution at scale. There have been many works assessing the robustness of watermarking to removal attacks, yet, watermark forging, the scenario when a watermark is stolen from genuine content and applied to malicious content, remains underexplored. In this work, we investigate watermark forging in the context of widely used post-hoc image watermarking. Our contributions are as follows. First, we introduce a preference model to assess whether an image is watermarked. The model is trained using a ranking loss on purely procedurally generated images without any need for real watermarks. Second, we demonstrate the model's capability to remove and forge watermarks by optimizing the input image through backpropagation. This technique requires only a single watermarked image and works without knowledge of the watermarking model, making our attack much simpler and more practical than attacks introduced in related work. Third, we evaluate our proposed method on a variety of post-hoc image watermarking models, demonstrating that our approach can effectively forge watermarks, questioning the security of current watermarking approaches. Our code and further resources are publicly available.

---

### FLORA: Unsupervised Knowledge Graph Alignment by Fuzzy Logic

Authors: Yiwen Peng, Thomas Bonald, Fabian M. Suchanek
Published: 2025-10-23
Categories: cs.AI, cs.DB
arXiv: https://arxiv.org/abs/2510.20467v1

Knowledge graph alignment is the task of matching equivalent entities (that is, instances and classes) and relations across two knowledge graphs. Most existing methods focus on pure entity-level alignment, computing the similarity of entities in some embedding space. They lack interpretable reasoning and need training data to work. In this paper, we propose FLORA, a simple yet effective method that (1) is unsupervised, i.e., does not require training data, (2) provides a holistic alignment for entities and relations iteratively, (3) is based on fuzzy logic and thus delivers interpretable results, (4) provably converges, (5) allows dangling entities, i.e., entities without a counterpart in the other KG, and (6) achieves state-of-the-art results on major benchmarks.

---

### Systematic Evaluation of Uncertainty Estimation Methods in Large Language Models

Authors: Christian Hobelsberger, Theresa Winner, Andreas Nawroth, Oliver Mitevski, Anna-Carolina Haensch
Published: 2025-10-23
Categories: cs.CL, stat.AP, stat.ME
arXiv: https://arxiv.org/abs/2510.20460v1

Large language models (LLMs) produce outputs with varying levels of uncertainty, and, just as often, varying levels of correctness; making their practical reliability far from guaranteed. To quantify this uncertainty, we systematically evaluate four approaches for confidence estimation in LLM outputs: VCE, MSP, Sample Consistency, and CoCoA (Vashurin et al., 2025). For the evaluation of the approaches, we conduct experiments on four question-answering tasks using a state-of-the-art open-source LLM. Our results show that each uncertainty metric captures a different facet of model confidence and that the hybrid CoCoA approach yields the best reliability overall, improving both calibration and discrimination of correct answers. We discuss the trade-offs of each method and provide recommendations for selecting uncertainty measures in LLM applications.

---

### Neural Reasoning for Robust Instance Retrieval in $\mathcal{SHOIQ}$

Authors: Louis Mozart Kamdem Teyou, Luke Friedrichs, N'Dah Jean Kouagou, Caglar Demir, Yasir Mahmood, Stefan Heindorf, Axel-Cyrille Ngonga Ngomo
Published: 2025-10-23
Categories: cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2510.20457v1

Concept learning exploits background knowledge in the form of description logic axioms to learn explainable classification models from knowledge bases. Despite recent breakthroughs in neuro-symbolic concept learning, most approaches still cannot be deployed on real-world knowledge bases. This is due to their use of description logic reasoners, which are not robust against inconsistencies nor erroneous data. We address this challenge by presenting a novel neural reasoner dubbed EBR. Our reasoner relies on embeddings to approximate the results of a symbolic reasoner. We show that EBR solely requires retrieving instances for atomic concepts and existential restrictions to retrieve or approximate the set of instances of any concept in the description logic $\mathcal{SHOIQ}$. In our experiments, we compare EBR with state-of-the-art reasoners. Our results suggest that EBR is robust against missing and erroneous data in contrast to existing reasoners.

---

### Capturing Intransitive Dominance in Tennis Forecasting: A Graph Neural Network Approach

Authors: Lawrence Clegg, John Cartlidge
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20454v2

Intransitive player dominance, where player A beats B, B beats C, but C beats A, is common in competitive tennis. Yet, there are few known attempts to incorporate it within forecasting methods. We address this problem with a graph neural network approach that explicitly models these intransitive relationships through temporal directed graphs, with players as nodes and their historical match outcomes as directed edges. Our model (65.7% accuracy, 0.214 Brier score) forecasts competitively with established rating systems such as Weighted Elo. Although it does not improve on the baseline in unconditional accuracy, a forecast-encompassing test shows that it carries complementary information. A combined forecast significantly outperforms Weighted Elo, and there is some indication that the gain grows more strongly on the intransitive matchups our model targets. A graph-based representation of player interactions thus captures a forecasting signal that transitive rating systems discard, even between players who share no common opponents.

---

### Symbolic Regression and Differentiable Fits in Beyond the Standard Model Physics

Authors: Shehu AbdusSalam, Steven Abel, Deaglan Bartlett, Miguel Crispim Romão
Published: 2025-10-23
Categories: hep-ph, astro-ph.CO, cs.AI, cs.LG, physics.comp-ph
arXiv: https://arxiv.org/abs/2510.20453v1

We demonstrate the efficacy of symbolic regression (SR) to probe models of particle physics Beyond the Standard Model (BSM), by considering the so-called Constrained Minimal Supersymmetric Standard Model (CMSSM). Like many incarnations of BSM physics this model has a number (four) of arbitrary parameters, which determine the experimental signals, and cosmological observables such as the dark matter relic density. We show that analysis of the phenomenology can be greatly accelerated by using symbolic expressions derived for the observables in terms of the input parameters. Here we focus on the Higgs mass, the cold dark matter relic density, and the contribution to the anomalous magnetic moment of the muon. We find that SR can produce remarkably accurate expressions. Using them we make global fits to derive the posterior probability densities of the CMSSM input parameters which are in good agreement with those performed using conventional methods. Moreover, we demonstrate a major advantage of SR which is the ability to make fits using differentiable methods rather than sampling methods. We also compare the method with neural network (NN) regression. SR produces more globally robust results, while NNs require data that is focussed on the promising regions in order to be equally performant.

---

### On Multiple Robustness of Proximal Dynamic Treatment Regimes

Authors: Yuanshan Gao, Yang Bai, Yifan Cui
Published: 2025-10-23
Categories: stat.ME, math.ST, stat.ML
arXiv: https://arxiv.org/abs/2510.20451v1

Dynamic treatment regimes are sequential decision rules that adapt treatment according to individual time-varying characteristics and outcomes to achieve optimal effects, with applications in precision medicine, personalized recommendations, and dynamic marketing. Estimating optimal dynamic treatment regimes via sequential randomized trials might face costly and ethical hurdles, often necessitating the use of historical observational data. In this work, we utilize proximal causal inference framework for learning optimal dynamic treatment regimes when the unconfoundedness assumption fails. Our contributions are four-fold: (i) we propose three nonparametric identification methods for optimal dynamic treatment regimes; (ii) we establish the semiparametric efficiency bound for the value function of a given regime; (iii) we propose a (K+1)-robust method for learning optimal dynamic treatment regimes, where K is the number of stages; (iv) as a by-product for marginal structural models, we establish identification and estimation of counterfactual means under a static regime. Numerical experiments validate the efficiency and multiple robustness of our proposed methods.

---

### LM-mixup: Text Data Augmentation via Language Model based Mixup

Authors: Zhijie Deng, Zhouan Shen, Ling Li, Yao Zhou, Zhaowei Zhu, Yanji He, Wei Wang, Jiaheng Wei
Published: 2025-10-23
Categories: cs.CL
arXiv: https://arxiv.org/abs/2510.20449v2

Instruction tuning is crucial for aligning Large Language Models (LLMs), yet the quality of instruction-following data varies significantly. While high-quality data is paramount, it is often scarce; conversely, abundant low-quality data is frequently discarded, leading to substantial information loss. Existing data augmentation methods struggle to augment this low-quality data effectively, and the evaluation of such techniques remains poorly defined. To address this, we formally define the task of Instruction Distillation: distilling multiple low-quality and redundant inputs into high-quality and coherent instruction-output pairs. Specifically, we introduce a comprehensive data construction pipeline to create MIXTURE, a 144K-sample dataset pairing low-quality or semantically redundant imperfect instruction clusters with their high-quality distillations. We then introduce LM-Mixup, by first performing supervised fine-tuning on MIXTURE and then optimizing it with reinforcement learning. This process uses three complementary reward signals: quality, semantic alignment, and format compliance, via Group Relative Policy Optimization (GRPO). We demonstrate that LM-Mixup effectively augments imperfect datasets: fine-tuning LLMs on its distilled data, which accounts for only about 3% of the entire dataset, not only surpasses full-dataset training but also competes with state-of-the-art high-quality data selection methods across multiple benchmarks. Our work establishes that low-quality data is a valuable resource when properly distilled and augmented with LM-Mixup, significantly enhancing the efficiency and performance of instruction-tuned LLMs.

---

### MolBridge: Atom-Level Joint Graph Refinement for Robust Drug-Drug Interaction Event Prediction

Authors: Xuan Lin, Aocheng Ding, Tengfei Ma, Hua Liang, Zhe Quan
Published: 2025-10-23
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2510.20448v2

Drug combinations offer therapeutic benefits but also carry the risk of adverse drug-drug interactions (DDIs), especially under complex molecular structures. Accurate DDI event prediction requires capturing fine-grained inter-drug relationships, which are critical for modeling metabolic mechanisms such as enzyme-mediated competition. However, existing approaches typically rely on isolated drug representations and fail to explicitly model atom-level cross-molecular interactions, limiting their effectiveness across diverse molecular complexities and DDI type distributions. To address these limitations, we propose MolBridge, a novel atom-level joint graph refinement framework for robust DDI event prediction. MolBridge constructs a joint graph that integrates atomic structures of drug pairs, enabling direct modeling of inter-drug associations. A central challenge in such joint graph settings is the potential loss of information caused by over-smoothing when modeling long-range atomic dependencies. To overcome this, we introduce a structure consistency module that iteratively refines node features while preserving the global structural context. This joint design allows MolBridge to effectively learn both local and global interaction outperforms state-of-the-art baselines, achieving superior performance across long-tail and inductive scenarios. patterns, yielding robust representations across both frequent and rare DDI types. Extensive experiments on two benchmark datasets show that MolBridge consistently. These results demonstrate the advantages of fine-grained graph refinement in improving the accuracy, robustness, and mechanistic interpretability of DDI event prediction.This work contributes to Web Mining and Content Analysis by developing graph-based methods for mining and analyzing drug-drug interaction networks.

---

### UniSE: A Unified Framework for Decoder-Only Autoregressive LM-Based Speech Enhancement

Authors: Haoyin Yan, Chengwei Liu, Shaofei Xue, Xiaotao Liang, Yinghao Liu, Yuxiang Kong, Zheng Xue
Published: 2025-10-23
Categories: cs.SD, cs.AI
arXiv: https://arxiv.org/abs/2510.20441v2

Neural audio codecs have largely promoted the application of language models (LMs) for speech applications. However, the effectiveness of autoregressive LM-based models in unifying speech enhancement (SE) tasks remains underexplored. In this work, we propose UniSE, a unified decoder-only LM-based framework to handle different SE tasks including speech restoration, target speaker extraction, and speech separation. Conditioned on input speech features, it autoregressively generates target discrete tokens, facilitating compatibility between distinct learning patterns of multiple tasks. To further optimize speech quality, we introduce a progressive reinforcement learning strategy with multiple assessment criteria. Experiments on several benchmarks show that UniSE achieves competitive performance compared to discriminative and generative baselines, demonstrating the capacity of LMs in unifying SE tasks. The code and demo are available at: https://github.com/alibaba/unified-audio/tree/main/QuarkAudio-UniSE.

---

### Explainable Benchmarking through the Lense of Concept Learning

Authors: Quannian Zhang, Michael Röder, Nikit Srivastava, N'Dah Jean Kouagou, Axel-Cyrille Ngonga Ngomo
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20439v1

Evaluating competing systems in a comparable way, i.e., benchmarking them, is an undeniable pillar of the scientific method. However, system performance is often summarized via a small number of metrics. The analysis of the evaluation details and the derivation of insights for further development or use remains a tedious manual task with often biased results. Thus, this paper argues for a new type of benchmarking, which is dubbed explainable benchmarking. The aim of explainable benchmarking approaches is to automatically generate explanations for the performance of systems in a benchmark. We provide a first instantiation of this paradigm for knowledge-graph-based question answering systems. We compute explanations by using a novel concept learning approach developed for large knowledge graphs called PruneCEL. Our evaluation shows that PruneCEL outperforms state-of-the-art concept learners on the task of explainable benchmarking by up to 0.55 points F1 measure. A task-driven user study with 41 participants shows that in 80\% of the cases, the majority of participants can accurately predict the behavior of a system based on our explanations. Our code and data are available at https://github.com/dice-group/PruneCEL/tree/K-cap2025

---

### Dynamic Weight Adjustment for Knowledge Distillation: Leveraging Vision Transformer for High-Accuracy Lung Cancer Detection and Real-Time Deployment

Authors: Saif Ur Rehman Khan, Muhammad Nabeel Asim, Sebastian Vollmer, Andreas Dengel
Published: 2025-10-23
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2510.20438v1

This paper presents the FuzzyDistillViT-MobileNet model, a novel approach for lung cancer (LC) classification, leveraging dynamic fuzzy logic-driven knowledge distillation (KD) to address uncertainty and complexity in disease diagnosis. Unlike traditional models that rely on static KD with fixed weights, our method dynamically adjusts the distillation weight using fuzzy logic, enabling the student model to focus on high-confidence regions while reducing attention to ambiguous areas. This dynamic adjustment improves the model ability to handle varying uncertainty levels across different regions of LC images. We employ the Vision Transformer (ViT-B32) as the instructor model, which effectively transfers knowledge to the student model, MobileNet, enhancing the student generalization capabilities. The training process is further optimized using a dynamic wait adjustment mechanism that adapts the training procedure for improved convergence and performance. To enhance image quality, we introduce pixel-level image fusion improvement techniques such as Gamma correction and Histogram Equalization. The processed images (Pix1 and Pix2) are fused using a wavelet-based fusion method to improve image resolution and feature preservation. This fusion method uses the wavedec2 function to standardize images to a 224x224 resolution, decompose them into multi-scale frequency components, and recursively average coefficients at each level for better feature representation. To address computational efficiency, Genetic Algorithm (GA) is used to select the most suitable pre-trained student model from a pool of 12 candidates, balancing model performance with computational cost. The model is evaluated on two datasets, including LC25000 histopathological images (99.16% accuracy) and IQOTH/NCCD CT-scan images (99.54% accuracy), demonstrating robustness across different imaging domains.

---

### Learning Decentralized Routing Policies via Graph Attention-based Multi-Agent Reinforcement Learning in Lunar Delay-Tolerant Networks

Authors: Federico Lozano-Cuadra, Beatriz Soret, Marc Sanchez Net, Abhishek Cauligi, Federico Rossi
Published: 2025-10-23
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2510.20436v1

We present a fully decentralized routing framework for multi-robot exploration missions operating under the constraints of a Lunar Delay-Tolerant Network (LDTN). In this setting, autonomous rovers must relay collected data to a lander under intermittent connectivity and unknown mobility patterns. We formulate the problem as a Partially Observable Markov Decision Problem (POMDP) and propose a Graph Attention-based Multi-Agent Reinforcement Learning (GAT-MARL) policy that performs Centralized Training, Decentralized Execution (CTDE). Our method relies only on local observations and does not require global topology updates or packet replication, unlike classical approaches such as shortest path and controlled flooding-based algorithms. Through Monte Carlo simulations in randomized exploration environments, GAT-MARL provides higher delivery rates, no duplications, and fewer packet losses, and is able to leverage short-term mobility forecasts; offering a scalable solution for future space robotic systems for planetary exploration, as demonstrated by successful generalization to larger rover teams.

---

### Partial Optimality in Cubic Correlation Clustering for General Graphs

Authors: David Stein, Bjoern Andres, Silvia Di Gregorio
Published: 2025-10-23
Categories: cs.DM, cs.LG
arXiv: https://arxiv.org/abs/2510.20431v1

The higher-order correlation clustering problem for a graph $G$ and costs associated with cliques of $G$ consists in finding a clustering of $G$ so as to minimize the sum of the costs of those cliques whose nodes all belong to the same cluster. To tackle this NP-hard problem in practice, local search heuristics have been proposed and studied in the context of applications. Here, we establish partial optimality conditions for cubic correlation clustering, i.e., for the special case of at most 3-cliques. We define and implement algorithms for deciding these conditions and examine their effectiveness numerically, on two data sets.

---

### An Empirical Study of Sample Selection Strategies for Large Language Model Repair

Authors: Xuran Li, Jingyi Wang
Published: 2025-10-23
Categories: cs.LG
arXiv: https://arxiv.org/abs/2510.20428v1

Large language models (LLMs) are increasingly deployed in real-world systems, yet they can produce toxic or biased outputs that undermine safety and trust. Post-hoc model repair provides a practical remedy, but the high cost of parameter updates motivates selective use of repair data. Despite extensive prior work on data selection for model training, it remains unclear which sampling criteria are most effective and efficient when applied specifically to behavioral repair of large generative models. Our study presents a systematic analysis of sample prioritization strategies for LLM repair. We evaluate five representative selection methods, including random sampling, K-Center, gradient-norm-based selection(GraNd), stratified coverage (CCS), and a Semantic-Aware Prioritized Sampling (SAPS) approach we proposed. Repair effectiveness and trade-offs are assessed through toxicity reduction, perplexity on WikiText-2 and LAMBADA, and three composite metrics: the Repair Proximity Score (RPS), the Overall Performance Score (OPS), and the Repair Efficiency Score (RES). Experimental results show that SAPS achieves the best balance between detoxification, utility preservation, and efficiency, delivering comparable or superior repair outcomes with substantially less data. Random sampling remains effective for large or robust models, while high-overhead methods such as CCS and GraNd provide limited benefit. The optimal data proportion depends on model scale and repair method, indicating that sample selection should be regarded as a tunable component of repair pipelines. Overall, these findings establish selection-based repair as an efficient and scalable paradigm for maintaining LLM reliability.

---