LLM Safety Evaluation

LLM: Large Language Model

Overview

The field tests whether language models refuse harmful requests, handle benign sensitive tasks helpfully, and remain safe when prompts, context, or tools change. Common approaches include fixed safety benchmarks, adversarial red-teaming and jailbreak testing, multi-turn agent simulations, domain-specific scenarios, and automated or human judging. Researchers increasingly measure more than refusal rates or attack success, including harm severity, omissions, false refusals, reasoning and tool-use failures, and effects on users. A central challenge is that results can shift with benchmark design, evaluation conditions, and judge choice, while stronger safeguards may reduce helpfulness and static tests may not predict behavior in deployment.

Latest papers

CardsList

277 papers25 papers in the last four weeks, up 9% on the four weeks before. 0.2% of all new papers.

  1. SEAM: Computer-Use Agents Know the Harm Between Screens, and Commit It Anyway

    Nov 17, 2026Yuxuan Zhou, Chen Xiong, Haitong Liu +5Computer-Use Agent SafetyLLM Safety Evaluation

  2. Can AI Judge You? An Empirical Audit of Deployment Readiness in LLM-assisted Hiring Systems

    Nov 15, 2026Spandan Roy, Subhabrata Majumdar, Soudeep DebAI-Assisted Decision MakingLLM Auditing

  3. The Critic Can Hurt: On the Safety of Critics in LLM-Guided AutoML Search

    Nov 14, 2026Gaoyuan Du, Xiaoyang Liu, Rex Zhou +4Automated Machine LearningLLM Safety Evaluation

  4. HandsOnAid: Building and Evaluating an Offline First-Aid Assistant for Emergency Response Scenarios

    Nov 10, 2026Rezwan Islam Salvi, Saiyma Sittul Muna, Mushfiqur Rahman MushfiqueMedical QAClinical Language Model Evaluation

  5. System Prompts Affect Response Bias More Than Discriminability in LLM Safety

    Nov 8, 2026Mohan W. Gupta, Sandy TanwisuthLLM Safety EvaluationRefusal Behavior in Language Models

  6. TrustMod-SM: A Multi-Axis Benchmark for Evaluating Trustworthiness of LLMs in Social Media Content Moderation

    Nov 7, 2026Aizan Zafar, Sreekantam Sai Venkata, Gathik Jindal +4LLM Safety BenchmarksLLM Safety Evaluation

  7. SafeStart: Mitigating the Overlooked Safety Cost of Active Expert Reduction in MoE LLM Serving

    Nov 4, 2026Erfan Baghaei Potraghloo, Saeid Shokoufa, Seyedarmin Azizi +4LLM SafetyMixture-of-Experts Language Models

  8. Widening Harm Evaluation Coverage with a Judge that Reads the World

    Nov 2, 2026Prabhpreet Singh Setia, Bhavesh Neekhra, Debayan GuptaAI Agent SafetyAI Safety Evaluation

  9. CoFiBreak: A Coarse-to-Fine Diversity Reward for Effective Multi-Turn Jailbreak

    Oct 29, 2026Jialin Song, Xiaodong Liu, Weiwei Yang +7LLM Red TeamingJailbreak Attacks

  10. RoleFUZZER: User-Behaviour Fuzzing for Role-Specification Adherence in Large Language Models

    Oct 28, 2026Kshitish Ghate, Michael Saxon, Tadayoshi Kohno +1LLM AuditingLLM Safety Evaluation

  11. BoardSafety: AI-Ready Data from Manufacturer Documentation for Development-Board Safety

    Oct 24, 2026Linjian Yang, Xinyan Wang, Kunpeng LiuAI Safety EvaluationLLM Safety Benchmarks

  12. Do AI-for-Science Agents Preserve Model Safety?

    Oct 23, 2026Michael Yew Chuan Lee, Manya Sriram, N. Benjamin ErichsonLLM Agent SafetyLLM Safety Evaluation

  13. Questioning Instrumental Convergence in RLVR: Checkpoint Evidence from OLMo-3

    Oct 22, 2026Barış Alparslan, Sandy Tanwisuth, Caleb PettersonRL Post-TrainingReinforcement Learning with Verifiable Rewards

  14. Diffusion Language Models for Perplexity-Aware Black-Box Attacks

    Oct 22, 2026Ali Al-Lawati, Elsayed Eshra, Dongwon Lee +1Adversarial Attacks on LLMsLLM Safety Evaluation

  15. From Clinical Judgment to Checkable Rules: Clinician-Led Guardrails for Mental-Health Chatbots

    Oct 22, 2026Chenhan Lyu, Kit Wislocki, Sarah Elizabeth Stoeckl +2LLM GuardrailsLLM Safety

  16. Beyond Binary Labels: Diagnostic Evaluation of LLM Safety with Prompt-Level Rubrics

    Oct 21, 2026Hongyu Li, Xinlong Huang, Zhan Yu +13LLM Safety EvaluationLLM Safety Benchmarks

  17. VEILED-SH: Vulnerability Expression through Indirect Language Evaluation Dataset (Self-Harm)

    Oct 19, 2026Leanne Tan, Keith Chua, Shaun Khoo +7LLM Safety BenchmarksLanguage Model Safety Evaluation

  18. A Calibrated Instrument for Detecting Disposition Drift in Language Models

    Oct 19, 2026Jerry KaplanLLM Safety MonitoringLLM Safety Evaluation

  19. Refuse the Harm, Complete the Task via Graph-of-Rules Planning for LLM-Enabled Robots

    Oct 17, 2026Ali Taheri, Alireza Taban, Qizhou Wang +1LLM Agent SafetyLarge Language Model-Based Robot Planning

  20. Risk and Refusal: Mechanisms of LLM Responses to Suicidal Ideation

    Oct 15, 2026Drishti Goel, Soorya Ram Shimgekar, Jeongah Lee +4LLM SafetyLLM Refusal Behavior

  21. Evaluation Awareness Beyond English: Do LLMs Act Differently When Tested in Indic Languages?

    Oct 15, 2026Trizal Garg, Sibayan Mitra, Arya Karanjkar +6Evaluation Awareness in Language ModelsLLM Safety Evaluation

  22. Functional Welfare for AI Safety

    Oct 15, 2026Rikhil Jhaveri, Jamie Johnson, Punya Syon Pandey +1LLM Safety EvaluationLLM Safety Monitoring