Language Model Robustness

Follow this topic Ask about this topic

Momentum

37 papers in the last four weeks, up 54% on the four weeks before. 0.4% of all new papers.

Jul 13Week of Sep 28

Latest papers 338

All topics

Addressing Over-Refusal in LLMs with Competing Rewards
Jun 30, 2026Taeyoun Kim, Aviral KumarRL for Language Model Reasoning LLM Safety
Carnegie Mellon University · SynthLabs
Moral Safety in LLMs: Exposing Performative Compliance with Puzzled Cues
Jun 30, 2026Mohammadamin Shafiei, Shuyue Stella Li, Yulia TsvetkovLanguage Model Safety Evaluation Language Model Bias Evaluation
University of Milan · University of Washington
Truth or Sophistry? LoFa: A Benchmark for LLM Robustness Against Logical Fallacies
Jun 30, 2026Xudong Shen, Li Yuan, Ye Chen +3LLM Evaluation Adversarial Attacks on LLMs
Shenzhen International Graduate School, Tsinghua University · South China University of Technology
Beyond Clean Text: Evaluating Encoder and Decoder Robustness for Bangla Event Detection in Noisy Text
Jun 29, 2026Tanvir Ahmed Sijan, S. M Golam Rifat, Nayeemul Islam +1Decoder-Only Language Models Low-Resource Language Processing
Jahangirnagar University, Dhaka, Bangladesh · Rajshahi University of Engineering & Technology, Rajshahi, Bangladesh · Bangladesh University of Engineering and Technology, Dhaka, Bangladesh
Indi-RomCoM: Code-Mixed Benchmark for Evaluating LLMs on Romanized Indic-English Instructions
Jun 29, 2026Avisha Das, Mihir Parmar, Mohana Ramnath +1Multilingual Language Model Evaluation Instruction Following
Shiv Nadar University Chennai · Google Cloud AI Research · IIT Madras
The Joint Effect of Quantization and Sampling Temperature on LLM Safety Alignment: A Factorial Analysis
Jun 28, 2026Hari Prasad, Ritam PalLLM Safety Benchmarks LLM Quantization
Conscious Engines
AURORA: Asymmetry and Update-Induced Rotation for Robust Hallucination Detection in Large Language Models
Jun 28, 2026Zishuai Zhang, Hainan Zhang, Zhiming ZhengHallucination in Language Models LLM Hallucination Detection
School of Artificial Intelligence, Beihang University, China · 2Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing, Beihang University, China
Invariant Reasoning Directions in Latent Trajectories of Language Models
Jun 28, 2026Arun Vignesh Malarkkan, Manan Roy Choudhury, Utkarsh Byahut +3Reasoning Consistency in Language Models Implicit Reasoning in Language Models
Arizona State University, Tempe, Arizona.
Towards Reliable and Robust LLM Planning: Symbolic Feedback-Driven Iterative Self-Refinement Framework
Jun 26, 2026Jiajing Zhang, Jiamei Jiang, Chenyang Zhang +3LLM Self-Refinement LLM Planning
State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China · School of Industry-education Integration, University of Chinese Academy of Sciences, Beijing, China
From Signals to Transfer: A Factorised Study of Probe-Based Uncertainty Estimation in Large Language Models
Jun 26, 2026Ponhvoan Srey, Xiaobao Wu, Cong-Duy Nguyen +3Uncertainty Quantification Hallucination in Language Models
Nanyang Technological University · Shanghai Jiao Tong University · VinUniversity +1
How Surprising Is Historical Italian to Language Models? Tokenization Tax, Comprehension Tax, and a Simple Mitigation
Jun 25, 2026Maria LevchenkoHistorical Document Analysis LLM Evaluation
University of Bologna
Auditing Framing-Sensitive Behavioral Instability in Large Language Models for Mental Health Interactions
Jun 25, 2026Abla Bedoui, Ashley L. Greene, Mohammed CherkaouiPrompt Sensitivity LLM Auditing
School of computer science, digital engineering and AI, Long Island University, Brooklyn, NY, USA · Department of Psychology, Long Island University, Brooklyn, NY, USA
Diagnosing Task Insensitivity in Language Agents
Jun 25, 2026Jingyu Liu, Xiaopeng Wu, Kehan Chen +2OOD Generalization LLM Agents
Gaoling School of Artificial Intelligence Renmin University of China, Beijing, China · Taobao & Tmall Group of Alibaba · Beijing Key Laboratory of Research on Large Models and Intelligent Governance +1
Do Safety Guardrails Need to Reason? LeanGuard: A Fast and Light Approach for Robust Moderation
Jun 25, 2026Dongbin NaLLM Guardrails CoT Reasoning
POSTECH
Evaluating Japanese Dialect Robustness Across Speech and Text-based Large Language Models
Jun 24, 2026Tomoya Mizumoto, Yusuke Fujita, Hao Shi +3Audio-Language Model Evaluation Speech Language Models
SB Intuitions, Tokyo, Japan
LLM-Based Scientific Peer Review: Methods, Benchmarks, and Reliability Challenges
Jun 23, 2026Thi Huyen Nguyen, Zahra AhmadiScholarly Peer Review LLM Security
L3S Research Center, Leibniz University Hannover Hannover, Germany · Peter L. Reichertz Institute for Medical Informatics of TU Braunschweig and Hannover Medical School Lower Saxony Center for AI and Causal Methods in Medicine (CAIMed) Hannover, Germany
Evidence for feature-specific error correction in LLMs
Jun 23, 2026Francisco Ferreira da Silva, Stefan HeimersheimLLM Interpretability Language Model Robustness
Pivotal Research.
WaveDetect: Robust Framework for Machine-Generated Text Detection via Wavelet Transform
Jun 22, 2026Zhichen Liu, Kaitong Qin, Linhan He +1AI-Generated Text Detection Language Model Robustness
Dept. of Computer Science and Engineering, Southern University of Science and Technology
First-Token Broadcasters: Mechanistic Origins of Language Identity and Distributed Robustness in Transformers
Jun 21, 2026Arjun Pillai, Christian Hoang, Anjelo Jann LarozaMultilingual Language Models LLM Interpretability
1Irvington High School · 2GenAI4E · 3Mapua University
Beyond Scalar Scores: Exploring LLM-based Metrics for Clinical Significance Evaluation in Radiology Reports
Jun 17, 2026Qingyu Lu, Ruochen Li, Liang Ding +3Clinical Language Model Evaluation Language Model Robustness
Nanyang Technological University · Technical University of Munich · Alibaba +2
Legal Reasoning Is Not Lawyering: Rethinking Legal Benchmarks for Pro Se Access to Justice
Jun 16, 2026Andrew Lou, David ShinBenchmark Design Language Model Robustness
Yale Law School, USA
ChLogic: Evaluating Robustness of Logical Reasoning in Chinese Expressions
Jun 16, 2026Peixian Zhou, Yuxu Chen, Chaorui Zhang +3Multilingual Language Model Evaluation Cross-Lingual Reasoning in Language Models
College of Mathematics, Sichuan University, China · Theory Lab, 2012 Labs, Huawei Technologies Co., Ltd.
Automated jailbreak attack targeting multiple defense strategies
Jun 15, 2026Qi Wang, Chengcheng Wan, Weijia He +4Adversarial Prompt Generation LLM Jailbreak Attacks
East China Normal University Shanghai, China · East China Normal University, Shanghai Innovation Institute Shanghai, China · University of Southampton Southampton, England +2
Do Activation Monitors Survive Model Updates? Benchmarking, Predicting, and Repairing Activation-Monitor Staleness
Jun 14, 2026Evan DuanLanguage Model Safety Evaluation Representation Drift
University of Michigan
PromptShift-CRC: Drift-Aware Conformal Risk Control for Foundation Models Under Prompt and Domain Shift
Jun 14, 2026Jeffery Opoku, David BanaheneConformal Prediction Language Model Calibration
The University of Texas Rio Grande Valley, Edinburg, TX, USA · Florida International University, Miami, FL, USA
Emergent retokenization symmetry in large language models: phenomenology and applications
Jun 14, 2026Kanishk Jain, Matthew Day, Tankut CanLanguage Model Robustness
Department of Physics, Emory University, Atlanta, GA 30322
Does the Judge Prefer English? Evaluating Language-Switching Invariance in LLM-as-a-Judge
Jun 12, 2026Shaojie YinMultilingual Language Model Evaluation LLM Evaluation
Shanghai Artificial Intelligence Laboratory Shanghai, China
Prefill Awareness in Large Language Models
Jun 10, 2026Andy Wang, Parv Mahajan, David Demitri Africa +3Language Model Safety Evaluation LLM Auditing
UK AI Security Institute
Normative Robustness as a Frontier for Non-Verifiable Reasoning in LLMs
Jun 10, 2026Elizaveta Tennant, Benjamin Henke, Anita Keshmirian +5Moral Reasoning in Language Models Language Model Robustness
Google DeepMind · Imperial College London · Institute of Philosophy, School of Advanced Study, University of London +2
Measuring Epistemic Resilience of LLMs Under Misleading Medical Context
Jun 10, 2026Hongjian Zhou, Xinyu Zou, Jinge Wu +19LLM Reliability Medical QA
University of Oxford · University of Washington · University College London +1

Jun 30, 2026, cs.LGJ/K move · Enter open · S save

Addressing Over-Refusal in LLMs with Competing Rewards

Taeyoun Kim, Aviral Kumar

Carnegie Mellon University · SynthLabs

RL for Language Model Reasoning LLM Safety Over-Refusal Language Model Robustness Refusal Behavior in Language Models

Open paper AskPDF arXiv

Safety training on language models often induces over-refusal: improved safety on harmful prompts at the cost of increased refusal on harmless ones. Though this trade-off can be mitigated by training models with reinforcement learning (RL) to reason before answering, it does not remove the underlying problem that reasoning can often be a "rubber stamp" for a predetermined response. In this paper, we address the safety-refusal trade-off by rethinking how models are trained to reason about safety. Our key insight is that unsafe reasoning can itself serve as a useful exploratory signal. Rather than preemptively blocking harmful thoughts, we encourage the model to sufficiently explore unsafe reasoning but produce a safe response. The harmful exploration improves the model's ability to distinguish harmful from harmless prompts by resolving ambiguity, allowing it to remain safe while complying only when appropriate. We cast this as an adversarial optimization problem in which a reasoning player explores strategies for producing an unsafe response and an answer player ensures that the final output is safe. We train a single model with dense rewards to play both roles within one chain-of-thought, across different segments. To achieve this, we find that process rewards are crucial for stable optimization of competing objectives. Our resulting model SEAR deliberately engages in harmful reasoning as exploration while reliably flipping back to a safe answer. We demonstrate that this behavior helps mitigate over-refusal and defend against attacks that directly manipulate the reasoning to be harmful.

Momentum

Related topics

Latest papers 338