arXiv cs.CL Computation and Language

Yuchen Zhang, Haralambos Mouratidis, Ravi Shekhar: Preserve to Expand: Source-Aware Adaptation and Model Merging for Multilingual LLM-Based ASR https://arxiv.org/abs/2610.35846 https://arxiv.org/pdf/2610.35846 https://arxiv.org/html/2610.35846

Mingyang Song, Mao Zheng, Zhiyuan Ma, Xinpeng Zhou: Understanding Jev-Style System One Models: Does Structure Reduce Selection Bias? https://arxiv.org/abs/2610.35826 https://arxiv.org/pdf/2610.35826 https://arxiv.org/html/2610.35826

Aidar Myrzakhan, Zhiqiang Shen: Learning When to Commit: Schedule Distillation for Fast Diffusion Language Models https://arxiv.org/abs/2610.35804 https://arxiv.org/pdf/2610.35804 https://arxiv.org/html/2610.35804

Tiasa Singha Roy, Manya Wadhwa, Greg Durrett: Quantifying Claim Smoothing by LLM Research Agents https://arxiv.org/abs/2610.35793 https://arxiv.org/pdf/2610.35793 https://arxiv.org/html/2610.35793

Feng, Liu, Yang, Xu, Song, Ma, Guo, Buractaon, White, Gao: SEMA-Auto: Sighted and Efficient Multi-Turn Attack with Autonomy https://arxiv.org/abs/2610.35743 https://arxiv.org/pdf/2610.35743 https://arxiv.org/html/2610.35743

Choi, Kuai, Li, Jiang, Zhang, Fan, Tu, Zhou: From Self-Confidence to Statistical Accuracy: Scaling Human-in-the-Loop Evaluation with Direct and LLM-Assisted Auditing https://arxiv.org/abs/2610.35698 https://arxiv.org/pdf/2610.35698 https://arxiv.org/html/2610.35698

Duc-Vinh Tran: Not One Belief: Answer Payloads, Persistence, and Dissociated Confidence Signals in Conversational Question Answering https://arxiv.org/abs/2610.35694 https://arxiv.org/pdf/2610.35694 https://arxiv.org/html/2610.35694

Hari Sitaula, Manish Sakhakarmy, Pujan Pant, Sagar Newpane: Adapting LLM-SC to Living Sacred Heritage: Domain-Specific Extraction, Semantic Networks and Human Evaluation on Pashupatinath Temp... https://arxiv.org/abs/2610.35685 https://arxiv.org/pdf/2610.35685 https://arxiv.org/html/2610.35685

Huang, Zeng, Wang, Zhu, Duan, Li, Xu, Wang, Tu, Xiong, Zang, Yan, Yang, Gu, Xu, Dong, Li: Evaluating DeepSeek V4.1 Flash-1017 for loss of control and harmful manipulation risks https://arxiv.org/abs/2610.35671 https://arxiv.org/pdf/2610.35671 https://arxiv.org/html/2610.35671

Haocheng Mei, Yuchuan Tian, Ye Yuan, Enhua Wu, Chao Xu, Allen Chen, Kai Han, Yunhe Wang: Improving JEV Entailment Reranking with Span Readouts and Selective Adaptation https://arxiv.org/abs/2610.35654 https://arxiv.org/pdf/2610.35654 https://arxiv.org/html/2610.35654

Joao P. A. Dantas: Offline Language-Based Support System for Tactical Decision-Making in Air Combat Operations https://arxiv.org/abs/2610.35635 https://arxiv.org/pdf/2610.35635 https://arxiv.org/html/2610.35635

Gao, Peng, Zhou, Zhu, Zeng, Zhou, Qin, Qiao, Ma, Shang, Jiang, Liao, Ren: Hierarchical Skill Evolution for Coding Agent https://arxiv.org/abs/2610.35616 https://arxiv.org/pdf/2610.35616 https://arxiv.org/html/2610.35616

Trystan Geoffre, Zheng Yuan, Thierry Geoffre: Making Reasoning Visible: Configuration-Aware Diagnostic Modeling for Adaptive Scaffolding in Structured Reasoning Tasks https://arxiv.org/abs/2610.35609 https://arxiv.org/pdf/2610.35609 https://arxiv.org/html/2610.35609

Yerim Hong, Sohyun Son, Heeseung Chung, Mujeen Sung: EXCISE: Locating and Excising Distracting Sentences in Medical Question Answering https://arxiv.org/abs/2610.35558 https://arxiv.org/pdf/2610.35558 https://arxiv.org/html/2610.35558

Feichen Song, Boyi Zeng, Yiqin Hao, Shixiang Song, He Li, Yizhou Yang, Zitong Wang, Yifan Liu, Ziwei He, Zhouhan Lin: LoopFold: Parallelizing Recurrent Depth in Looped Transformers https://arxiv.org/abs/2610.35556 https://arxiv.org/pdf/2610.35556 https://arxiv.org/html/2610.35556

Feichen Song, Boyi Zeng, He Li, Shixiang Song, Yiqin Hao, Zitong Wang, Yizhou Yang, Yifan Liu, Ziwei He, Zhouhan Lin: StateLoop: Persistent State for Looped Transformers https://arxiv.org/abs/2610.35544 https://arxiv.org/pdf/2610.35544 https://arxiv.org/html/2610.35544

Michael A. Lepori, Laila Johnston, Aalok Sathe, Zhuonan Yang, Roman Feiman, Ellie Pavlick: An Algorithmic Account of Few-Shot Concept Induction in Language Models https://arxiv.org/abs/2610.35533 https://arxiv.org/pdf/2610.35533 https://arxiv.org/html/2610.35533

Bingrui Jin, Baihan Li, Zitong Li, Kunyao Lan, Shimeng Li, Kenny Q. Zhu, Mengyue Wu: MATCHA: A Chinese Multi-Disorder Mental Health Dataset with Cross-Lingual Analyses on Social Media https://arxiv.org/abs/2610.35512 https://arxiv.org/pdf/2610.35512 https://arxiv.org/html/2610.35512

Li, Qiu, Wu, Li, Wang, Lyu, Guo, Zhang, Xu, Lu, Shen: Beyond Words Alone: Disentangling Distributed Evidence for Visually Grounded Cultural Translation https://arxiv.org/abs/2610.35482 https://arxiv.org/pdf/2610.35482 https://arxiv.org/html/2610.35482

Liu, Yang, Wang, Zhu, Ren, Zhang, Wang, Chen: MANTIS: Multi-Axis News Transformations for Instance-Wise Stress Testing and Worst-Case Training in Fake News Detection https://arxiv.org/abs/2610.35475 https://arxiv.org/pdf/2610.35475 https://arxiv.org/html/2610.35475

Sally Dong, Luke Thorburn, Andrew Konya, Michiel Bakker: A Contextual Measure of Ambiguity in Prescriptive Text https://arxiv.org/abs/2610.35461 https://arxiv.org/pdf/2610.35461 https://arxiv.org/html/2610.35461

Artem Saloev, Nicolas Ballier, Erin Pacquetet: Probing the phon* representations of the Mimi codebooks https://arxiv.org/abs/2610.35432 https://arxiv.org/pdf/2610.35432 https://arxiv.org/html/2610.35432

Timo Oksanen: INVAR-Agri: Detecting Discrepancies in Multilingual Operator Manuals of Agricultural Implements with a Fine-Tuned LLM https://arxiv.org/abs/2610.35423 https://arxiv.org/pdf/2610.35423 https://arxiv.org/html/2610.35423

\v{S}\'arka Uramov\'a, Petar Kirilov Petrov, Josef Jon, Ond\v{r}ej Bojar, Michal Nov\'ak: ImgMT: Wikipedia-based Dataset For Evaluating Text-in-Image Translation https://arxiv.org/abs/2610.35416 https://arxiv.org/pdf/2610.35416 https://arxiv.org/html/2610.35416

Adam Hollowell: The Geometry of Emotion Representations Across Scale, Depth, and Instruction Tuning https://arxiv.org/abs/2610.35411 https://arxiv.org/pdf/2610.35411 https://arxiv.org/html/2610.35411

Nikita A. Klimin (HSE University, Moscow, Russia): Agreement Between Reconstructed and Native GDELT GKG Records: A Field-Level Baseline https://arxiv.org/abs/2610.35358 https://arxiv.org/pdf/2610.35358 https://arxiv.org/html/2610.35358

Ngoc Trinh Hung Nguyen, Alonso Silva, Armen Aghasaryan, Mehwish Alam: Constrained Decoding: Harmful or Beneficial? Revisiting Its Impact on LLM Performance https://arxiv.org/abs/2610.35328 https://arxiv.org/pdf/2610.35328 https://arxiv.org/html/2610.35328

Zikai Zhou, Jiacheng Sang, Yilin Xu, Yufei Jin, Monica S. Lam: Did the Model Use That? Counterfactual Training Boosts Self-Attribution in Language Models https://arxiv.org/abs/2610.35295 https://arxiv.org/pdf/2610.35295 https://arxiv.org/html/2610.35295

Hao Shen, Ruichen Qiu, Lihong Zhi, Xiao-Shan Gao: MechAlign: Evaluating Semantic Alignment at the Project Level in Mathematical Formalization https://arxiv.org/abs/2610.35269 https://arxiv.org/pdf/2610.35269 https://arxiv.org/html/2610.35269

Malik Marmonier, Beno\^it Sagot, Rachel Bawden: The Lay of the Land in Idiom-Translation Quality Estimation https://arxiv.org/abs/2610.35261 https://arxiv.org/pdf/2610.35261 https://arxiv.org/html/2610.35261

Vasu, Joshi, Christmann, Sheth, Binkyte, Chowdhury, Kerkouche, Wang, Fritz: Style over Substance? Linguistic Bias in LLM-based Peer Review Copy title https://arxiv.org/abs/2610.35254 https://arxiv.org/pdf/2610.35254 https://arxiv.org/html/2610.35254

Chen, Pan, Liang, Li, Ma, Shen, Wang, Dahlmeier, Zhang, Zhang: DTCA: Denoising Token Credit Assignment for Reinforcement Learning in Diffusion Language Models https://arxiv.org/abs/2610.35201 https://arxiv.org/pdf/2610.35201 https://arxiv.org/html/2610.35201

Jiang, Liu, Wang, Overney, Zhang, Luk, Yang, Roy, Kabbara: Lost in Translation: Investigating Systematic Discrepancies between Parallel English and Chinese Names of American Chinese Restaurants https://arxiv.org/abs/2610.35191 https://arxiv.org/pdf/2610.35191 https://arxiv.org/html/2610.35191

Elena Senger, Yuri Campbell, Rob van der Goot, Barbara Plank: High Scores, Hidden Failures: Evaluating Codebook-Based LLM Content Analysis https://arxiv.org/abs/2610.35179 https://arxiv.org/pdf/2610.35179 https://arxiv.org/html/2610.35179

Uzor, Lemofouet, Anyanwu, Abdullahi, Siro, Abdulmumin, Yimam, Muhammad: Do Language Models Understand Igbo Culture? Beyond Multiple Choice https://arxiv.org/abs/2610.35168 https://arxiv.org/pdf/2610.35168 https://arxiv.org/html/2610.35168

Qianchi Zhang, Hainan Zhang, Liang Pang, Hongwei Zheng, Zhiming Zheng: CalibraRL: Confidence-Calibrated Reinforcement Learning for Mitigating Hallucinations in RAG https://arxiv.org/abs/2610.35142 https://arxiv.org/pdf/2610.35142 https://arxiv.org/html/2610.35142

Zenith Biswas, Andrew Thomas Dyer: Quality over Quantity: Rethinking Data and Model Assumptions for Bengali Coreference Resolution https://arxiv.org/abs/2610.35114 https://arxiv.org/pdf/2610.35114 https://arxiv.org/html/2610.35114

Niyar R Barman, Omar Naim, Nicholas Asher: Rethinking Evaluation for Principled Generalization Using a Kripkean Challenge to Meaning as Use https://arxiv.org/abs/2610.35111 https://arxiv.org/pdf/2610.35111 https://arxiv.org/html/2610.35111

Ventirozos, Armitage, Oshagh, Appleby, Shardlow: Useful Diversity or Hallucinated Variety? Persona-Conditioned Synthetic Query Generation for E-Commerce Search https://arxiv.org/abs/2610.35102 https://arxiv.org/pdf/2610.35102 https://arxiv.org/html/2610.35102

Ivana Crescenzi, Arianna Muti, Maarten Sap, Tommaso Caselli, Debora Nozza: RAGing Against Misogyny: Improving Implicit Misogyny Explanations https://arxiv.org/abs/2610.35094 https://arxiv.org/pdf/2610.35094 https://arxiv.org/html/2610.35094

Jinhu Qi, Minda Hu, Wentao Zhang, Zixing Song, Irwin King: Ask Before You Answer: A Question-Driven Evidence Reasoning Architecture (QERA) for In-Context Learning https://arxiv.org/abs/2610.35083 https://arxiv.org/pdf/2610.35083 https://arxiv.org/html/2610.35083

Yujia Zheng, Tianhao Li, Shangzhi Xu, Yuekai Huang, Ziyou Jiang, Zhaoyang Li, Yuyao Ge, Junran Peng, Mingyang Li: CoSA: The Conflict of Skills Attack in Real-World LLM Agent Ecosystems https://arxiv.org/abs/2610.35082 https://arxiv.org/pdf/2610.35082 https://arxiv.org/html/2610.35082

Jingwen Zhu, Keshu Wu, Steven T. Parker, Bin Ran, David A. Noyce: From Questions to Queries to Insights: A Schema-Grounded Agentic Workflow for Natural-Language Crash Data Analysis https://arxiv.org/abs/2610.35054 https://arxiv.org/pdf/2610.35054 https://arxiv.org/html/2610.35054

Haotian Tan, Hiroki Ouchi, Sakriani Sakti: SimPOTrans: Self-Adaptation for Offline-to-Simultaneous Speech Translation via Reinforcement Learning https://arxiv.org/abs/2610.35042 https://arxiv.org/pdf/2610.35042 https://arxiv.org/html/2610.35042

Bastian Bunzeck, Francesca Padovani: Synthetic Child-Adult Interactions Are Not Developmentally Plausible https://arxiv.org/abs/2610.35029 https://arxiv.org/pdf/2610.35029 https://arxiv.org/html/2610.35029

Joonyong Park, Jerry Li: When the Persona Contradicts the Line: Diagnosing Plan-to-Synthesis Failures in Chain-of-Thought TTS https://arxiv.org/abs/2610.35008 https://arxiv.org/pdf/2610.35008 https://arxiv.org/html/2610.35008

Junho Kim, CheHyun Lee, Donghun Kang, Youngho Cho, Dongwook Lee, Heeseung Kim, Sungroh Yoon: Diagnosing and Mitigating Long-Session Degradation in Full-Duplex Speech Models https://arxiv.org/abs/2610.34991 https://arxiv.org/pdf/2610.34991 https://arxiv.org/html/2610.34991

Xue, Somani, Singh, Mu, Wang, Liu, Khatun, Kwon, Ma, Ma, He, Yu, Li: AskGLP1: A Knowledge Graph LLM Agent for Exploring GLP-1 Experience from Online Patient Narratives https://arxiv.org/abs/2610.34987 https://arxiv.org/pdf/2610.34987 https://arxiv.org/html/2610.34987

Sanyal, Chen, Kuriyan, Tak, Kothandaraman, Mehra, Kimball, Tumanov: SVD-Attention: Efficiently Analyzing Attention Blocks in Transformers https://arxiv.org/abs/2610.34971 https://arxiv.org/pdf/2610.34971 https://arxiv.org/html/2610.34971

Shan Yang: What Outcome RL Buys: Finishing, Reliability, Rarely Reach https://arxiv.org/abs/2610.34894 https://arxiv.org/pdf/2610.34894 https://arxiv.org/html/2610.34894