arXiv cs.CL Computation and Language

Yu, Lee, Lee, Agarwal, Manolache, Kang, Zaharia, Mahoney, Keutzer, Kim, Gholami: GEPA-Curator: Curated Learning for Self Improving Agents https://arxiv.org/abs/2610.37651 https://arxiv.org/pdf/2610.37651 https://arxiv.org/html/2610.37651

Zhen Yang, Wen Wang, Xander Xu, Ying-Cong Chen: Can LLMs Uncover Hidden Physical Laws through Free Exploration? https://arxiv.org/abs/2610.37649 https://arxiv.org/pdf/2610.37649 https://arxiv.org/html/2610.37649

Hee Seung Hwang, Sanghyuk Chun, Xindi Wu, Olga Russakovsky, Zhiwei Deng: Multi-State Linear Attention with Generalized Second-Order Routing https://arxiv.org/abs/2610.37645 https://arxiv.org/pdf/2610.37645 https://arxiv.org/html/2610.37645

Mingyu Zhao, Shuhang Lin, Ligong Han, Xiaojing Yu, Zhenting Wang, Yifei Pang, Yang Zhou, Dimitris N. Metaxas: Metacognitive Decoding: Efficient Inference for Diffusion Language Agents https://arxiv.org/abs/2610.37642 https://arxiv.org/pdf/2610.37642 https://arxiv.org/html/2610.37642

Devon Yanitski, Chris Westbury: semanticfa: An R Package for Response-Free Semantic Factor Analysis of Psychometric Scales https://arxiv.org/abs/2610.37494 https://arxiv.org/pdf/2610.37494 https://arxiv.org/html/2610.37494

Trent R. Northen, Mingxun Wang: Teach reality, not ethics: a drop-in continued-pretraining corpus installs coupled human-biosphere-AI welfare dispositions across model families https://arxiv.org/abs/2610.37484 https://arxiv.org/pdf/2610.37484 https://arxiv.org/html/2610.37484

Md Sakib Ul Islam, S M Mehedi Zaman: What Carries Forward? Pretraining Disclosure Across Language Model Releases https://arxiv.org/abs/2610.37465 https://arxiv.org/pdf/2610.37465 https://arxiv.org/html/2610.37465

Xinying Qian, Ying Zhang, Yu Zhao, Xubo Liu, Baohang Zhou, Jeff Z. Pan: ETAMem: Decoupling and Aligning Heterogeneous Memory for Long-Term Conversational Agents https://arxiv.org/abs/2610.37443 https://arxiv.org/pdf/2610.37443 https://arxiv.org/html/2610.37443

Haokun Wen, Xuemeng Song, Mong-Li Lee, Wynne Hsu: Same Verdict, Different Readers: Personalizing Fact-Checking Explanations from Relative Preferences https://arxiv.org/abs/2610.37380 https://arxiv.org/pdf/2610.37380 https://arxiv.org/html/2610.37380

Emily Chen, Michelle Menachery, Himanshu Kumar, Rahul Nenavath, Amirfarrokh Iranitalab: UCVE: Graph Structured Retrieval for Compliance Validation https://arxiv.org/abs/2610.37323 https://arxiv.org/pdf/2610.37323 https://arxiv.org/html/2610.37323

Tim Menzner, Jochen L. Leidner: Multi-Label Agreement: A Flexible Framework for Inter-Annotator Reliability https://arxiv.org/abs/2610.37292 https://arxiv.org/pdf/2610.37292 https://arxiv.org/html/2610.37292

Ahmed M. Fetouh, Mariam Labib, Nsrin Ashraf, Omer Dawood, Hamada Nayel: REGLAT at ArGuard Shared Tasks: An Ensemble-Based Model Integrating NER for Harmful Content Detection in Arabic https://arxiv.org/abs/2610.37265 https://arxiv.org/pdf/2610.37265 https://arxiv.org/html/2610.37265

Lemofouet, Uzor, Anyanwu, Kapsa, Aliyu, Yimam, Muhammad: Aligning Large Language Models for Safety in African Languages: A Case Study of Hausa https://arxiv.org/abs/2610.37245 https://arxiv.org/pdf/2610.37245 https://arxiv.org/html/2610.37245

Haozheng Liu, Zhaoqian Xue, Zhongyin Wang, Junting Mu, Zihui Ma, Shuya Feng, Lingyao Li: Decision Models as Judges for Hallucination Detection https://arxiv.org/abs/2610.37229 https://arxiv.org/pdf/2610.37229 https://arxiv.org/html/2610.37229

Nabeelah Faumi, Deshan Sumanathilaka: What Survives Distillation? A Mechanistic Interpretability Study of GPT-2 and DistilGPT-2 https://arxiv.org/abs/2610.37157 https://arxiv.org/pdf/2610.37157 https://arxiv.org/html/2610.37157

Sinanta, Al Kautsar, Izzah, Nguyen, Chau, Lim, Koto: Text Generation in Low-Resource Languages Using Human-Curated Vocabularies https://arxiv.org/abs/2610.37105 https://arxiv.org/pdf/2610.37105 https://arxiv.org/html/2610.37105

Rong Huang, Yi Xiong, Yuan-Yuan Cheng, Xiao-Ming Fu: Hierarchical Low-Rank Reparameterization for Parameter-Efficient https://arxiv.org/abs/2610.37068 https://arxiv.org/pdf/2610.37068 https://arxiv.org/html/2610.37068

Manan Uppadhyay, Himanshu Beniwal, Prashant Kodali, Sunayana Sitaram: The State and Fate of Multilingual, Contextual Evaluation in the NLP World https://arxiv.org/abs/2610.36990 https://arxiv.org/pdf/2610.36990 https://arxiv.org/html/2610.36990

XinYu Piao, Myunghoon Ryu, Joongheon Kim, Jong-Kook Kim: Self-Keyed In-Parameter Memorization for External Memory-Free Dialogue Recall in Frozen Language Models https://arxiv.org/abs/2610.36948 https://arxiv.org/pdf/2610.36948 https://arxiv.org/html/2610.36948

Qiyong Zhong, Jiajie Su, Ming Yang, Li Zhang: JevRec: Linguistic Encoding with Jev-like Decision Modeling for LLM-based Generative Recommendation https://arxiv.org/abs/2610.36937 https://arxiv.org/pdf/2610.36937 https://arxiv.org/html/2610.36937

Saifullah Saifullah, David Dembinsky, Stefan Agne, Andreas Dengel, Sheraz Ahmed: DocXEval: A Systematic Evaluation of Explainability in Multimodal Document Understanding https://arxiv.org/abs/2610.36929 https://arxiv.org/pdf/2610.36929 https://arxiv.org/html/2610.36929

Mingyu Zhang, Zixu Li, Zhiheng Fu, Zhiwei Chen, Jinhe Bi, Yupeng Hu: TrawMem: Reframing Agent Memory as Task-Oriented Role-Adaptive Workspaces https://arxiv.org/abs/2610.36918 https://arxiv.org/pdf/2610.36918 https://arxiv.org/html/2610.36918

Parv Khatri, Harshul Jain, Mehak Kapoor, Tanya Sah: Do We Need an LLM to Know Why Sources Conflict? Decision Models for Knowledge-Conflict Typing in RAG https://arxiv.org/abs/2610.36915 https://arxiv.org/pdf/2610.36915 https://arxiv.org/html/2610.36915

Andoni Sudupe, Inma Hernaez, Ander Arriandiaga, Christoforos Souganidis, Ibon Saratxaga, Eva Navas: Developing a Speech Language Model for Basque https://arxiv.org/abs/2610.36911 https://arxiv.org/pdf/2610.36911 https://arxiv.org/html/2610.36911

Martinelli, Lorenzo, Garg, Wang, Orlando, Piriyatamwong, Fraiberger: Nothing Left Behind: Finding Evidence from a Large Randomized Evaluation Corpus https://arxiv.org/abs/2610.36910 https://arxiv.org/pdf/2610.36910 https://arxiv.org/html/2610.36910

Zhang, Ma, Liao, Zhou, Wang, Zhao, Zhang, Si, Tang, Li, Liu, Bian: Words or Deeds? How LLMs Predict Users' Decisions in Human-Agent Dialogue https://arxiv.org/abs/2610.36813 https://arxiv.org/pdf/2610.36813 https://arxiv.org/html/2610.36813

Runze Jiang, Yan Wang, Boran Jin, Bingyu Zhu, Longtao Huang: Weakness Localization Is Not Enough: State-Aware Recursive Knowledge Improvement for Large Language Models https://arxiv.org/abs/2610.36811 https://arxiv.org/pdf/2610.36811 https://arxiv.org/html/2610.36811

Gunhee Cho, Yun-Gyung Cheong: Personality Latents: Conditioning Frozen Language Models on Individuals from Their Own Writing https://arxiv.org/abs/2610.36805 https://arxiv.org/pdf/2610.36805 https://arxiv.org/html/2610.36805

Xiang Ding: Language Models Keep Abandoned Analyses: Residue and Adaptation after Garden-Path Sentences https://arxiv.org/abs/2610.36797 https://arxiv.org/pdf/2610.36797 https://arxiv.org/html/2610.36797

Huang, Tan, Miao, Gong, Wang, Zhou, Li, Yan: DocSkill: Skill-Integrated Agentic Reasoning for Multilingual Multi-Hop Document Question Answering https://arxiv.org/abs/2610.36759 https://arxiv.org/pdf/2610.36759 https://arxiv.org/html/2610.36759

Jingpu Yang, Tianhao Li, Fengxian Ji, Yufeng Wang, Shuxin Zheng, Xiaoqing Zhang: CAMR: Compatibility-Aware Memory Replay for Continual Learning in Large Language Models https://arxiv.org/abs/2610.36734 https://arxiv.org/pdf/2610.36734 https://arxiv.org/html/2610.36734

Xiaochen Chang: A Harness Reducing over-Defensive Writing in Agents by Isolating Objections https://arxiv.org/abs/2610.36717 https://arxiv.org/pdf/2610.36717 https://arxiv.org/html/2610.36717

Pragna Mallappa: Is Token Confidence Just Chain Length? A Signal-Processing Audit of Reasoning Traces https://arxiv.org/abs/2610.36709 https://arxiv.org/pdf/2610.36709 https://arxiv.org/html/2610.36709

Steven Y. Feng, Michael C. Frank: Specialization, Diversity, and Scaling in Child-Scale Language Models https://arxiv.org/abs/2610.36679 https://arxiv.org/pdf/2610.36679 https://arxiv.org/html/2610.36679

Senthil Rajasekaran, Stylianos Loukas Vasileiou: A Quantitative Approach to Asch Conformity in Frontier Large Language Models https://arxiv.org/abs/2610.36668 https://arxiv.org/pdf/2610.36668 https://arxiv.org/html/2610.36668

Satomu Tomisaki, Yosuke Mikami, Amane Watahiki, Hitomi Yanaka: Japanese Discourse Structure Analysis Based on Segmented Discourse Representation Theory Using LLMs https://arxiv.org/abs/2610.36667 https://arxiv.org/pdf/2610.36667 https://arxiv.org/html/2610.36667

Mifu Matsumoto, Yuki Nakanishi, Tomoya Hashiguchi, Hiroaki Ohshima: Structured Maps for Landmark-Based Route Instructions with LLMs https://arxiv.org/abs/2610.36582 https://arxiv.org/pdf/2610.36582 https://arxiv.org/html/2610.36582

Qian, Du, Peng, Li, Peng, Xue, Chen, Wang, Yu: Agentic Data Sampling Framework for Multilingual ASR via Linguistic-Aware Error-Driven Optimization https://arxiv.org/abs/2610.36538 https://arxiv.org/pdf/2610.36538 https://arxiv.org/html/2610.36538

Zhiheng Qian, Junhao Du, Jing Peng, Hanqi Li, Kunyang Peng, Feng Xue, Kai Yu: ETC-Bench: A Multilingual Benchmark for Evaluation-Side Transcript Canonicalization https://arxiv.org/abs/2610.36527 https://arxiv.org/pdf/2610.36527 https://arxiv.org/html/2610.36527

Jipeng Li, Pramod Khargonekar, Yanning Shen: Causal Localization of Evidence Effects in Transformers https://arxiv.org/abs/2610.36524 https://arxiv.org/pdf/2610.36524 https://arxiv.org/html/2610.36524

Wenyuan Li, Guang Li, Dongyuan Li, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama: Benchmarking Task Takeover Across Agent Handoffs https://arxiv.org/abs/2610.36523 https://arxiv.org/pdf/2610.36523 https://arxiv.org/html/2610.36523

Miri Liu, Xiayu Zhao, Encheng Xie, ChengXiang Zhai: Do AI Scientists Know What's New? https://arxiv.org/abs/2610.36511 https://arxiv.org/pdf/2610.36511 https://arxiv.org/html/2610.36511

Jixuan Leng, Si Si, Fei Wang, Cho-Jui Hsieh, Jiong Zhang, Hsiang-Fu Yu, Inderjit S. Dhillon: CONCUR: Concordant Auxiliary Rewards for Multi-Task, Multi-Reward Reinforcement Learning https://arxiv.org/abs/2610.36437 https://arxiv.org/pdf/2610.36437 https://arxiv.org/html/2610.36437

Alexander Nemecek, Yuqiao Xu, Joseph Falkenburg, Erman Ayday: No Tokenizer, Still a Gap: Cross-Lingual Fairness of Homoglyph and Invisible-Character Watermarking https://arxiv.org/abs/2610.36432 https://arxiv.org/pdf/2610.36432 https://arxiv.org/html/2610.36432

Ali Khairallah, Arkaitz Zubiaga: AK_QMUL at ArGuard Shared Tasks: A Single Fine-Tuned JAIS for Arabic Harmful-Prompt Detection https://arxiv.org/abs/2610.36360 https://arxiv.org/pdf/2610.36360 https://arxiv.org/html/2610.36360

Chen, Liu, Clifford, Moradipari, Chen, Han, Ammar, Wei: Detect Before You Disambiguate: Perturbation-Based Input Ambiguity Detection in LLMs https://arxiv.org/abs/2610.36353 https://arxiv.org/pdf/2610.36353 https://arxiv.org/html/2610.36353

Hafsteinn Einarsson: Beyond IceBERT: Open-Data ModernBERT Encoders for Icelandic from 12M to 1B Parameters https://arxiv.org/abs/2610.36230 https://arxiv.org/pdf/2610.36230 https://arxiv.org/html/2610.36230

Devansh Sharma, Faizul Rakib Sayem, Shahana Ibrahim: LLM Hallucination Detection via Low-rank Multilinear Decomposition https://arxiv.org/abs/2610.36191 https://arxiv.org/pdf/2610.36191 https://arxiv.org/html/2610.36191

Michael A. Robinette, Christina Barsky, Amanda D. Clark, Helen Flannery, Juniper Lovato: Frankly Speaking: The Company Election Integrity Keeps in Congressional Communication https://arxiv.org/abs/2610.36187 https://arxiv.org/pdf/2610.36187 https://arxiv.org/html/2610.36187

Sanjid Hasan, A H M Fuad, Risalat Labib, Bayazid Hasan Simon: Lipi-Ghor: A Large-Scale, Diarized, Multi-Domain Bengali Speech Corpus and a Scalable Pipeline for Building It https://arxiv.org/abs/2610.36184 https://arxiv.org/pdf/2610.36184 https://arxiv.org/html/2610.36184