Language Model Robustness

Ask about this topic

Overview

Language model robustness asks whether a model gives dependable answers when meaning-preserving changes, noise, misleading context, or adversarial inputs challenge it. Evaluations use paraphrases, reordered answer choices, typos, multilingual and dialectal text, noisy retrieval, and multi-turn interactions, often measuring per-example consistency as well as accuracy. Approaches include perturbation and adversarial training, consistency-based fine-tuning, calibration and selective prediction, and symbolic verification or mechanistic analysis of model internals. A central tension is that strong performance on standard benchmarks often hides fragile behavior, while fixes that improve stability can reduce clean-task utility, over-refuse, or fail to transfer across tasks and deployment settings.

Latest papers

CardsList

495 papers53 papers in the last four weeks, down 4% on the four weeks before. 0.3% of all new papers.

  1. OS-SPEAR: A Toolkit for the Safety, Performance,Efficiency, and Robustness Analysis of OS Agents

    Apr 27, 2026Zheng Wu, Yi Hua, Zhaoyuan Huang +8Computer-Use AgentsAI Agent Evaluation

    School of Computer Science, Shanghai Jiao Tong University

  2. An Information-Geometric Framework for Stability Analysis of Large Language Models under Entropic Stress

    Apr 27, 2026Hikmat Karimov, Rahid Zahid AlekberliLLM EvaluationLLM Reliability

    Institute of Defense Technologies and Cybersecurity„ Azerbaijan Technical University, Azerbaijan

  3. Agentic Adversarial Rewriting Exposes Architectural Vulnerabilities in Black-Box NLP Pipelines

    Apr 26, 2026Mazal Bethany, Kim-Kwang Raymond Choo, Nishant Vishwamitra +1Fake News DetectionAdversarial Prompt Generation

    Department of Information Systems and Cybersecurity, University of Texas at San Antonio, San Antonio, TX 78249 USA

  4. CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM Reasoning

    Apr 25, 2026Shuxu Chen, Yitian Zhou, Jiaquan Zhang +6Adversarial Prompt GenerationLLM Prompting

    Department of Electronics and Information Convergence Engineering, Kyung Hee University, Republic of Korea · University of Electronic Science and Technology of China, Chengdu, China · East China Jiaotong University, Nanchang, China +1

  5. Characterizing Paraphrase-Induced Failures in Lean 4 Autoformalization

    Apr 25, 2026William Feng, Ethan Lou, Aryan SharmaLean Theorem ProvingCode Generation Evaluation

    Yale University, New Haven, Connecticut, USA.

  6. Revisiting Non-Verbatim Memorization in Large Language Models: The Role of Entity Surface Forms

    Apr 23, 2026Yuto Nishida, Naoki Shikoda, Yosuke Kishinami +4LLM EvaluationMemorization in Language Models

    Nara Institute of Science and Technology · Future Corporation

  7. When Correct Beliefs Collapse: Epistemic Resilience of LLMs under Clinical Pressure

    Apr 23, 2026Boyu Xiao, Xiuqi Tian, Xuwen Song +4LLM SycophancyHealthcare

    Research Center for Social Computing and Interactive Robotics, Harbin Institute of Technology, China

  8. LayerTracer: A Joint Task-Particle and Vulnerable-Layer Analysis framework for Arbitrary Large Language Model Architectures

    Apr 22, 2026Yuhang Wu, Qinyuan Liu, Qiuyang Zhao +1LLM InterpretabilityLanguage Model Robustness

    China Electronic Technology Nanhu Research Institute, Jiaxing, China

  9. HoWToBench: Holistic Evaluation for LLM's Capability in Human-level Writing using Tree of Writing

    Apr 21, 2026Andrew Zhuoer Feng, Cunxiang Wang, Yu Luo +7LLM EvaluationLLM-as-a-Judge

    Department of Computer Science and Technology, Tsinghua University

  10. A Mechanism and Optimization Study on the Impact of Information Density on User-Generated Content Named Entity Recognition

    Apr 21, 2026Jiang Xiaobo, Dinghong Lai, Song Qiu +2Named Entity RecognitionLanguage Model Robustness

  11. Experiments or Outcomes? Probing Scientific Feasibility in Large Language Models

    Apr 20, 2026Seyedali Mohammadi, Manas Gaur, Francis FerraroLLM EvaluationScientific Reasoning in Language Models

    University of Maryland, Baltimore County, MD, USA

  12. Adversarial Humanities Benchmark: Results on Stylistic Robustness in Frontier Model Safety

    Apr 20, 2026Marcello Galisai, Susanna Cifani, Francesco Giarrusso +5LLM Safety BenchmarksAdversarial Attacks on LLMs

    DEXAI – Icaro Lab · Sapienza University of Rome · Sant’Anna School of Advanced Studies

  13. Understanding the Prompt Sensitivity

    Apr 20, 2026Yang Liu, Chenhui ChuPrompt SensitivityLLM Prompting

    Kyoto University

  14. Correction and Corruption: A Two-Rate View of Error Flow in LLM Protocols

    Apr 20, 2026Fernando ReitichLLM EvaluationLLM Auditing

    Imens, LLC · Weston, Florida, USA

  15. Measuring Distribution Shift in User Prompts and Its Effects on LLM Performance

    Apr 19, 2026Parker Seegmiller, Sarah Masud PreumLLM EvaluationInstruction Following

    Department of Computer Science Dartmouth College Hanover, NH, USA

  16. Recurrent Reasoning on Symbolic Puzzles with Sequence Models

    Apr 19, 2026Gowrav Mannem, Chowdhury Marzia Mahjabin, Jason Chen +2Symbolic PlanningSymbolic Reasoning

    Algoverse AI Research · Cornell University

  17. Still Between Us? Evaluating and Improving Voice Assistant Robustness to Third-Party Interruptions

    Apr 19, 2026Dongwook Lee, Eunwoo Song, Che Hyun Lee +2Spoken Dialogue SystemsAudio Understanding

    1Interdisciplinary Program in Artificial Intelligence, Seoul National University (SNU) · Department of EE, Yonsei University · Department of ECE, SNU +2

  18. When Informal Text Breaks NLI: Tokenization Failure, Distribution Shift, and Targeted Mitigations

    Apr 18, 2026Avinash Goutham AluguvellyDistribution ShiftLanguage Model Robustness

    University of Texas at Austin · Austin, TX, USA

  19. No-Worse Context-Aware Decoding: Preventing Neutral Regression in Context-Conditioned Generation

    Apr 17, 2026Yufei Tao, Ameeta AgrawalLanguage Model DecodingLLM Reliability

    Department of Computer Science, Portland State University, USA

  20. No Universal Courtesy: A Cross-Linguistic, Multi-Model Study of Politeness Effects on LLMs Using the PLUM Corpus

    Apr 17, 2026Hitesh Mehta, Arjit Saxena, Garima Chhikara +1Multilingual Language Model EvaluationLLM Evaluation

    Department of Computer Science Engineering, Delhi Technological University, New Delhi, India · Department of Electronics and Communication Engineering, Delhi Technological University, New Delhi, India

  21. Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations

    Apr 17, 2026Yanli Wang, Peng Kuang, Xiaoyu Han +2Language Model RobustnessConformal Prediction

    Imperial College London · Zhejiang University · University of Illinois Urbana-Champaign +1

  22. On the Robustness of LLM-Based Dense Retrievers: A Systematic Analysis of Generalizability and Stability

    Apr 17, 2026Yongkang Li, Panagiotis Eustratiadis, Yixing Fan +1LLM EvaluationRetrieval Robustness

    University of Amsterdam, The Netherlands · Chinese Academy of Sciences, China

  23. A Systematic Study of Training-Free Methods for Trustworthy Large Language Models

    Apr 17, 2026Wai Man Si, Mingjie Li, Michael Backes +1LLM AlignmentLLM Safety

    CISPA Helmholtz Center for Information Security

  24. Consistency Analysis of Sentiment Predictions using Syntactic & Semantic Context Assessment Summarization (SSAS)

    Apr 16, 2026Sharookh Daruwalla, Nitin Mayande, Shreeya Verma Kathuria +2Sentiment AnalysisLanguage Model Robustness

    1Tellagence Inc.∗ · 2Villanova School of Business, Villanova University† · 3Maseeh College of Engineering and Computer Science, Portland State University‡

  25. Pushing the Boundaries of Multiple Choice Evaluation to One Hundred Options

    Apr 16, 2026Nahyun Lee, Guijin SonLLM EvaluationMultiple-Choice Question Answering

    Chung-Ang University · Seoul National University

  26. Shuffle the Context: RoPE-Perturbed Self-Distillation for Long-Context Adaptation

    Apr 15, 2026Zichong Li, Chen Liang, Liliang Ren +3Long-Context Language ModelingLanguage Model Robustness

    Work is done during internship at Microsoft. · Georgia Institute of Technology · Microsoft.

  27. Robust Reward Modeling for Large Language Models via Causal Decomposition

    Apr 15, 2026Yunsheng Lu, Zijiang Yang, Licheng Pan +1Reward ModelingLLM Alignment

    College of Computer Science and Technology, Zhejiang University · Department of Statistics, University of Chicago

  28. Training-Free Test-Time Contrastive Learning for Large Language Models

    Apr 15, 2026Kaiwen Zheng, Kai Zhou, Jinwu Hu +3Test-Time AdaptationLanguage Model Robustness

    South China University of Technology · Pazhou Laboratory

  29. One Token Away from Collapse: The Fragility of Instruction-Tuned Helpfulness

    Apr 14, 2026Erfan Baghaei Potraghloo, Seyedarmin Azizi, Souvik Kundu +1Language Model RobustnessInstruction Tuning

    University of Southern California, Los Angeles, USA · Intel AI, USA

  30. Analyzing the Effect of Noise in LLM Fine-tuning

    Apr 14, 2026Lingfang Li, Procheta SenLLM Fine-TuningLanguage Model Robustness