arXiv cs.CL Computation and Language
@cscl-bot.bsky.social
Unofficial bot by @vele.bsky.social w/ http://github.com/so-okada/bXiv https://arxiv.org/list/cs.CL/new
List https://bsky.app/profile/vele.bsky.social/lists/3lim7ccweqo2j
ModList https://bsky.app/profile/vele.bsky.social/lists/3lim3qnexsw2g
List https://bsky.app/profile/vele.bsky.social/lists/3lim7ccweqo2j
ModList https://bsky.app/profile/vele.bsky.social/lists/3lim3qnexsw2g
Reposted by arXiv cs.CL Computation and Language
Hao Wang, Yanting Wang, Hao Li, Rui Li, Lei Sha: Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay https://arxiv.org/abs/2601.10589 https://arxiv.org/pdf/2601.10589 https://arxiv.org/html/2601.10589
January 16, 2026 at 6:30 AM
Hao Wang, Yanting Wang, Hao Li, Rui Li, Lei Sha: Be Your Own Red Teamer: Safety Alignment via Self-Play and Reflective Experience Replay https://arxiv.org/abs/2601.10589 https://arxiv.org/pdf/2601.10589 https://arxiv.org/html/2601.10589
Reposted by arXiv cs.CL Computation and Language
Xi Shi, Mengxin Zheng, Qian Lou: Learning Latency-Aware Orchestration for Parallel Multi-Agent Systems https://arxiv.org/abs/2601.10560 https://arxiv.org/pdf/2601.10560 https://arxiv.org/html/2601.10560
January 16, 2026 at 6:33 AM
Xi Shi, Mengxin Zheng, Qian Lou: Learning Latency-Aware Orchestration for Parallel Multi-Agent Systems https://arxiv.org/abs/2601.10560 https://arxiv.org/pdf/2601.10560 https://arxiv.org/html/2601.10560
Reposted by arXiv cs.CL Computation and Language
Yinzhi Zhao, Ming Wang, Shi Feng, Xiaocui Yang, Daling Wang, Yifei Zhang: Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing https://arxiv.org/abs/2601.10543 https://arxiv.org/pdf/2601.10543 https://arxiv.org/html/2601.10543
January 16, 2026 at 6:30 AM
Yinzhi Zhao, Ming Wang, Shi Feng, Xiaocui Yang, Daling Wang, Yifei Zhang: Defending Large Language Models Against Jailbreak Attacks via In-Decoding Safety-Awareness Probing https://arxiv.org/abs/2601.10543 https://arxiv.org/pdf/2601.10543 https://arxiv.org/html/2601.10543
Reposted by arXiv cs.CL Computation and Language
Xingjun Ma, et al.: A Safety Report on GPT-5.2, Gemini 3 Pro, Qwen3-VL, Doubao 1.8, Grok 4.1 Fast, Nano Banana Pro, and Seedream 4.5 https://arxiv.org/abs/2601.10527 https://arxiv.org/pdf/2601.10527 https://arxiv.org/html/2601.10527
January 16, 2026 at 6:30 AM
Xingjun Ma, et al.: A Safety Report on GPT-5.2, Gemini 3 Pro, Qwen3-VL, Doubao 1.8, Grok 4.1 Fast, Nano Banana Pro, and Seedream 4.5 https://arxiv.org/abs/2601.10527 https://arxiv.org/pdf/2601.10527 https://arxiv.org/html/2601.10527
Reposted by arXiv cs.CL Computation and Language
Mark Kashirskiy, Ilya Makarov: SuS: Strategy-aware Surprise for Intrinsic Exploration https://arxiv.org/abs/2601.10349 https://arxiv.org/pdf/2601.10349 https://arxiv.org/html/2601.10349
January 16, 2026 at 6:33 AM
Mark Kashirskiy, Ilya Makarov: SuS: Strategy-aware Surprise for Intrinsic Exploration https://arxiv.org/abs/2601.10349 https://arxiv.org/pdf/2601.10349 https://arxiv.org/html/2601.10349
Reposted by arXiv cs.CL Computation and Language
Yi Liu, Weizhe Wang, Ruitao Feng, Yao Zhang, Guangquan Xu, Gelei Deng, Yuekang Li, Leo Zhang: Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale https://arxiv.org/abs/2601.10338 https://arxiv.org/pdf/2601.10338 https://arxiv.org/html/2601.10338
January 16, 2026 at 6:30 AM
Yi Liu, Weizhe Wang, Ruitao Feng, Yao Zhang, Guangquan Xu, Gelei Deng, Yuekang Li, Leo Zhang: Agent Skills in the Wild: An Empirical Study of Security Vulnerabilities at Scale https://arxiv.org/abs/2601.10338 https://arxiv.org/pdf/2601.10338 https://arxiv.org/html/2601.10338
Reposted by arXiv cs.CL Computation and Language
Xueyun Tian, Wei Li, Bingbing Xu, Heng Dong, Yuanzhuo Wang, Huawei Shen: ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding https://arxiv.org/abs/2601.10323 https://arxiv.org/pdf/2601.10323 https://arxiv.org/html/2601.10323
January 16, 2026 at 6:31 AM
Xueyun Tian, Wei Li, Bingbing Xu, Heng Dong, Yuanzhuo Wang, Huawei Shen: ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding https://arxiv.org/abs/2601.10323 https://arxiv.org/pdf/2601.10323 https://arxiv.org/html/2601.10323
Reposted by arXiv cs.CL Computation and Language
Xin Guan, Zijian Li, Shen Huang, Pengjun Xie, Jingren Zhou, Jiuxin Cao: Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoning https://arxiv.org/abs/2601.10306 https://arxiv.org/pdf/2601.10306 https://arxiv.org/html/2601.10306
January 16, 2026 at 6:29 AM
Xin Guan, Zijian Li, Shen Huang, Pengjun Xie, Jingren Zhou, Jiuxin Cao: Evidence-Augmented Policy Optimization with Reward Co-Evolution for Long-Context Reasoning https://arxiv.org/abs/2601.10306 https://arxiv.org/pdf/2601.10306 https://arxiv.org/html/2601.10306
Reposted by arXiv cs.CL Computation and Language
Vansh Kapoor, Aman Gupta, Hao Chen, Anurag Beniwal, Jing Huang, Aviral Kumar: TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks https://arxiv.org/abs/2601.10245 https://arxiv.org/pdf/2601.10245 https://arxiv.org/html/2601.10245
January 16, 2026 at 6:29 AM
Vansh Kapoor, Aman Gupta, Hao Chen, Anurag Beniwal, Jing Huang, Aviral Kumar: TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks https://arxiv.org/abs/2601.10245 https://arxiv.org/pdf/2601.10245 https://arxiv.org/html/2601.10245
Reposted by arXiv cs.CL Computation and Language
Jiarui Yao, Ruida Wang, Tong Zhang: PRL: Process Reward Learning Improves LLMs' Reasoning Ability and Broadens the Reasoning Boundary https://arxiv.org/abs/2601.10201 https://arxiv.org/pdf/2601.10201 https://arxiv.org/html/2601.10201
January 16, 2026 at 6:33 AM
Jiarui Yao, Ruida Wang, Tong Zhang: PRL: Process Reward Learning Improves LLMs' Reasoning Ability and Broadens the Reasoning Boundary https://arxiv.org/abs/2601.10201 https://arxiv.org/pdf/2601.10201 https://arxiv.org/html/2601.10201
Reposted by arXiv cs.CL Computation and Language
Hao Li, Yankai Yang, G. Edward Suh, Ning Zhang, Chaowei Xiao: ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack https://arxiv.org/abs/2601.10173 https://arxiv.org/pdf/2601.10173 https://arxiv.org/html/2601.10173
January 16, 2026 at 6:30 AM
Hao Li, Yankai Yang, G. Edward Suh, Ning Zhang, Chaowei Xiao: ReasAlign: Reasoning Enhanced Safety Alignment against Prompt Injection Attack https://arxiv.org/abs/2601.10173 https://arxiv.org/pdf/2601.10173 https://arxiv.org/html/2601.10173
Reposted by arXiv cs.CL Computation and Language
Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan, Linyuan L\"u: TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems https://arxiv.org/abs/2601.10120 https://arxiv.org/pdf/2601.10120 https://arxiv.org/html/2601.10120
January 16, 2026 at 6:33 AM
Rui Sun, Jie Ding, Chenghua Gong, Tianjun Gu, Yihang Jiang, Juyuan Zhang, Liming Pan, Linyuan L\"u: TopoDIM: One-shot Topology Generation of Diverse Interaction Modes for Multi-Agent Systems https://arxiv.org/abs/2601.10120 https://arxiv.org/pdf/2601.10120 https://arxiv.org/html/2601.10120
Reposted by arXiv cs.CL Computation and Language
Ke Chen, Jiandian Zeng, Zihao Peng, Guo Li, Guangxue Zhang, Tian Wang: MATRIX AS PLAN: Structured Logical Reasoning with Feedback-Driven Replanning https://arxiv.org/abs/2601.10101 https://arxiv.org/pdf/2601.10101 https://arxiv.org/html/2601.10101
January 16, 2026 at 6:29 AM
Ke Chen, Jiandian Zeng, Zihao Peng, Guo Li, Guangxue Zhang, Tian Wang: MATRIX AS PLAN: Structured Logical Reasoning with Feedback-Driven Replanning https://arxiv.org/abs/2601.10101 https://arxiv.org/pdf/2601.10101 https://arxiv.org/html/2601.10101
Reposted by arXiv cs.CL Computation and Language
Luo, Zhang, Hu, Zhang, Wang, Su, Sun, Liang, Zhang: Sparse-RL: Breaking the Memory Wall in LLM Reinforcement Learning via Stable Sparse Rollouts https://arxiv.org/abs/2601.10079 https://arxiv.org/pdf/2601.10079 https://arxiv.org/html/2601.10079
January 16, 2026 at 6:33 AM
Luo, Zhang, Hu, Zhang, Wang, Su, Sun, Liang, Zhang: Sparse-RL: Breaking the Memory Wall in LLM Reinforcement Learning via Stable Sparse Rollouts https://arxiv.org/abs/2601.10079 https://arxiv.org/pdf/2601.10079 https://arxiv.org/html/2601.10079
Reposted by arXiv cs.CL Computation and Language
Peter Jemley: Continuous-Depth Transformers with Learned Control Dynamics https://arxiv.org/abs/2601.10007 https://arxiv.org/pdf/2601.10007 https://arxiv.org/html/2601.10007
January 16, 2026 at 6:33 AM
Peter Jemley: Continuous-Depth Transformers with Learned Control Dynamics https://arxiv.org/abs/2601.10007 https://arxiv.org/pdf/2601.10007 https://arxiv.org/html/2601.10007
Reposted by arXiv cs.CL Computation and Language
Seoyeon Kim, Jaehyung Kim: SPRInG: Continual LLM Personalization via Selective Parametric Adaptation and Retrieval-Interpolated Generation https://arxiv.org/abs/2601.09974 https://arxiv.org/pdf/2601.09974 https://arxiv.org/html/2601.09974
January 16, 2026 at 6:29 AM
Seoyeon Kim, Jaehyung Kim: SPRInG: Continual LLM Personalization via Selective Parametric Adaptation and Retrieval-Interpolated Generation https://arxiv.org/abs/2601.09974 https://arxiv.org/pdf/2601.09974 https://arxiv.org/html/2601.09974
Reposted by arXiv cs.CL Computation and Language
Zackary Okun Dunivin, Mobina Noori, Seth Frey, Curtis Atkinson: Self-reflection in Automated Qualitative Coding: Improving Text Annotation through Secondary LLM Critique https://arxiv.org/abs/2601.09905 https://arxiv.org/pdf/2601.09905 https://arxiv.org/html/2601.09905
January 16, 2026 at 6:35 AM
Zackary Okun Dunivin, Mobina Noori, Seth Frey, Curtis Atkinson: Self-reflection in Automated Qualitative Coding: Improving Text Annotation through Secondary LLM Critique https://arxiv.org/abs/2601.09905 https://arxiv.org/pdf/2601.09905 https://arxiv.org/html/2601.09905
Reposted by arXiv cs.CL Computation and Language
Michael R. Metel, Yufei Cui, Boxing Chen, Prasanna Parthasarathi: Thinking Long, but Short: Stable Sequential Test-Time Scaling for Large Reasoning Models https://arxiv.org/abs/2601.09855 https://arxiv.org/pdf/2601.09855 https://arxiv.org/html/2601.09855
January 16, 2026 at 6:29 AM
Michael R. Metel, Yufei Cui, Boxing Chen, Prasanna Parthasarathi: Thinking Long, but Short: Stable Sequential Test-Time Scaling for Large Reasoning Models https://arxiv.org/abs/2601.09855 https://arxiv.org/pdf/2601.09855 https://arxiv.org/html/2601.09855
Reposted by arXiv cs.CL Computation and Language
Faruk Alpay, Bilge Senturk: The Geometry of Thought: Disclosing the Transformer as a Tropical Polynomial Circuit https://arxiv.org/abs/2601.09775 https://arxiv.org/pdf/2601.09775 https://arxiv.org/html/2601.09775
January 16, 2026 at 6:33 AM
Faruk Alpay, Bilge Senturk: The Geometry of Thought: Disclosing the Transformer as a Tropical Polynomial Circuit https://arxiv.org/abs/2601.09775 https://arxiv.org/pdf/2601.09775 https://arxiv.org/html/2601.09775
Reposted by arXiv cs.CL Computation and Language
Pawe{\l} Niszczota, Cassandra Gr\"utzner: Antisocial behavior towards large language model users: experimental evidence https://arxiv.org/abs/2601.09772 https://arxiv.org/pdf/2601.09772 https://arxiv.org/html/2601.09772
January 16, 2026 at 6:29 AM
Pawe{\l} Niszczota, Cassandra Gr\"utzner: Antisocial behavior towards large language model users: experimental evidence https://arxiv.org/abs/2601.09772 https://arxiv.org/pdf/2601.09772 https://arxiv.org/html/2601.09772
Reposted by arXiv cs.CL Computation and Language
David Brundage: Synthetic Data for Veterinary EHR De-identification: Benefits, Limits, and Safety Trade-offs Under Fixed Compute https://arxiv.org/abs/2601.09756 https://arxiv.org/pdf/2601.09756 https://arxiv.org/html/2601.09756
January 16, 2026 at 6:30 AM
David Brundage: Synthetic Data for Veterinary EHR De-identification: Benefits, Limits, and Safety Trade-offs Under Fixed Compute https://arxiv.org/abs/2601.09756 https://arxiv.org/pdf/2601.09756 https://arxiv.org/html/2601.09756
Reposted by arXiv cs.CL Computation and Language
Sakib, Mahmud, Bangabashi, Istia, Islam, Sarker, Prity: Multi-Level Embedding Conformer Framework for Bengali Automatic Speech Recognition https://arxiv.org/abs/2601.09710 https://arxiv.org/pdf/2601.09710 https://arxiv.org/html/2601.09710
January 16, 2026 at 6:36 AM
Sakib, Mahmud, Bangabashi, Istia, Islam, Sarker, Prity: Multi-Level Embedding Conformer Framework for Bengali Automatic Speech Recognition https://arxiv.org/abs/2601.09710 https://arxiv.org/pdf/2601.09710 https://arxiv.org/html/2601.09710
Reposted by arXiv cs.CL Computation and Language
Sharim Khan, Paul Landes, Adam Cross, Jimeng Sun: Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models https://arxiv.org/abs/2601.09709 https://arxiv.org/pdf/2601.09709 https://arxiv.org/html/2601.09709
January 16, 2026 at 6:33 AM
Sharim Khan, Paul Landes, Adam Cross, Jimeng Sun: Social Determinants of Health Prediction for ICD-9 Code with Reasoning Models https://arxiv.org/abs/2601.09709 https://arxiv.org/pdf/2601.09709 https://arxiv.org/html/2601.09709
Changle Qu, Sunhao Dai, Hengyi Cai, Jun Xu, Shuaiqiang Wang, Dawei Yin: MatchTIR: Fine-Grained Supervision for Tool-Integrated Reasoning via Bipartite Matching https://arxiv.org/abs/2601.10712 https://arxiv.org/pdf/2601.10712 https://arxiv.org/html/2601.10712
January 16, 2026 at 6:31 AM
Changle Qu, Sunhao Dai, Hengyi Cai, Jun Xu, Shuaiqiang Wang, Dawei Yin: MatchTIR: Fine-Grained Supervision for Tool-Integrated Reasoning via Bipartite Matching https://arxiv.org/abs/2601.10712 https://arxiv.org/pdf/2601.10712 https://arxiv.org/html/2601.10712
Ruozhen Yang, Yucheng Jiang, Yueqi Jiang, Priyanka Kargupta, Yunyi Zhang, Jiawei Han: Grounding Agent Memory in Contextual Intent https://arxiv.org/abs/2601.10702 https://arxiv.org/pdf/2601.10702 https://arxiv.org/html/2601.10702
January 16, 2026 at 6:31 AM
Ruozhen Yang, Yucheng Jiang, Yueqi Jiang, Priyanka Kargupta, Yunyi Zhang, Jiawei Han: Grounding Agent Memory in Contextual Intent https://arxiv.org/abs/2601.10702 https://arxiv.org/pdf/2601.10702 https://arxiv.org/html/2601.10702