LLM Safety Benchmarks

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

23 papers in the last four weeks, up 44% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 141

All topics

Single Canonical Prompts Underestimate LLM Safety's Surface-Form Sensitivity
Aug 1, 2026Yongxi Zhou, Junwei Yao, Yuanzhe Liu +4LLM Safety Benchmarks Prompt Sensitivity
1Northeastern University, Massachusetts, USA · 2Georgia Institute of Technology, Georgia, USA · University of Southern California, California, USA
InfoOps Bench: A live information operations safety benchmark
Jul 30, 2026Dorian Quelle, Lisa-Maria Neudert, Jonathan Bright +1LLM Safety Benchmarks Language Model Safety Evaluation
Prosody-driven Jailbreaks in Audio LLMs: A Controlled Study and Mechanistic Analysis
Jul 29, 2026Jiachen Qian, Junyu LiAudio-Language Model Evaluation LLM Safety Benchmarks
City University of Hong Kong · City University of Hong Kong (Dongguan)
Inspect India Evals: An Open Benchmarking Framework for Evaluating Large Language Models in the Indian Linguistic and Cultural Context
Jul 28, 2026Abhishek Kumar Singh, Shrey Nag, Sachita +2LLM Safety Benchmarks Multilingual Language Model Evaluation
GTBIT¹ · TIET² · IGDTUW³ +2
Poster: Rethinking Security in LLM Code Generation through Real-World Risk Scenarios
Jul 25, 2026Lixun Ma, Ruolong Ma, Bei Wang +4LLM Safety Benchmarks LLM Security
Zhejiang University, Hangzhou, China · University College London, London, United Kingdom · Artificial Intelligence Institute, CAICT, Beijing, China +1
Beyond Shapley: An Influence-Based Data Auditing Pipeline for LLM Alignment and Evaluation
Jul 24, 2026Yunting Song, Matthew Watson, Peter Grabowski +1LLM Safety Benchmarks LLM Auditing
Google Mountain View, CA 94043
What AI Red-Team Evaluations Can and Cannot Prove
Jul 23, 2026Bandana KaurLLM Safety Benchmarks LLM Red Teaming
APIsec Research Labs
SciHazard: A Benchmark for Measuring Scientific Safety Risks with Decomposed Harm Scoring
Jul 21, 2026Chunxiao Li, Yuan Xiong, Lijun Li +4LLM Safety Benchmarks AI Agent Safety Benchmarks
Shanghai Artificial Intelligence Laboratory
JailMeter: An Evidence-Based Evaluation Framework for Jailbreak Attacks on Large Language Models
Jul 20, 2026Qingjia Huang, Jingyu Zhang, Jianguo Wu +6LLM Safety Benchmarks LLM Jailbreak Attacks
Institute of Information Engineering, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Renmin University of China +2
Silent Alarm: A J-Space Protocol for Comparing Danger Recognition Across Models and Quantization Levels
Jul 14, 2026Roman Prosvirnin, Victor Minchenkov, Alexey Soldatov +1LLM Safety Benchmarks Language Model Safety Evaluation
HSE University
AMT-X: Phase-Structured Multi-Turn Red-Teaming with Checklist-Gated Evaluation
Jul 13, 2026Yi Ting Shen, Kentaroh Toyoda, Alex LeungLLM Safety Benchmarks Adversarial Attacks on LLMs
Vulcan Research, AIFT, Singapore
Minionese: Comprehensive Benchmark and Mechanistic Study of Multilingual LLM Safety
Jul 11, 2026Chigozirim Ifebi, Brent Kong, Ayushi MehrotraLLM Safety Benchmarks Multilingual Language Model Evaluation
California Institute of Technology.
Trivial Prompt Reframing Bypasses Safety Guardrails in Googleś MedGemma-4B
Jul 9, 2026Avi-ad Avraam BuskilaLLM Safety Benchmarks Language Model Safety Evaluation
Department of Information Science and Applied Artificial Intelligence, Bar-Ilan University, Ramat Gan, Israel
Seduced by the Narrative: Assessing Rule Adherence in Semi-Open Textual Sandboxes
Jul 2, 2026Weiying Chen, Junlong Shen, Zhanyuan Guo +1LLM Safety Benchmarks Adversarial Prompt Generation
University of Alberta, Edmonton, AB, Canada · N-Dice Association · Qilu University of Technology, Shandong, China
Out-of-Distribution Generalization of Risk Aversion in Language Models
Jul 2, 2026Kristina Zhang, Junior Chinomso Okoroafor, Benjamin Maltbie +3LLM Safety Benchmarks OOD Generalization
Massachusetts Institute of Technology, Cambridge, MA, USA.
HaloGuard 1.0: An Open Weights Constitutional Classifier for Multilingual AI Safety
Jul 2, 2026Navaneeth Sangameswaran, Preetham S, Ashmiya LeninLLM Safety Benchmarks Language Model Safety Evaluation
Astroware AI, Delaware, USA · Karunya Institute of Technology and Sciences, Coimbatore, India
OpenSafeIntent: Evaluating Intent-Calibrated Safe Completion Across Dual-Use Prompt Sets
Jul 2, 2026Rheeya Uppaal, Seungwoo Lyu, Selina Sung +1LLM Safety Benchmarks Language Model Safety Evaluation
Department of Computer Sciences University of Wisconsin-Madison · Department of CSE Korea University
AIriskEval-edu: New Dataset for Risk Assessment in AI-mediated K-12 Educational Explanations
Jul 2, 2026Javier Irigoyen, Roberto Daza, Francisco Jurado +5LLM Safety Benchmarks LLM Auditing
BiometricsAI, Universidad Aut´onoma de Madrid (UAM), Spain · GHIA, Universidad Aut´onoma de Madrid (UAM), Spain · Universidad de Las Palmas de Gran Canaria (ULPGC), Spain
Adversarial Pragmatics for AI Safety Evaluation: A Benchmark for Instruction Conflict, Embedded Commands, and Policy Ambiguity
Jul 1, 2026Brett ReynoldsLLM Safety Benchmarks Language Model Safety Evaluation
Humber Polytechnic & University of Toronto
The Moving Target: A Longitudinal Audit of Trustworthiness Drift Across Twelve Checkpoints of Open-Source Chat LLMs
Jul 1, 2026Zhichao Fan, Yanhang Li, Zexin Zhuang +2LLM Safety Benchmarks LLM Evaluation
University of Illinois Urbana-Champaign · Northeastern University · Southern Methodist University +2
Security--Fidelity Tradeoffs: The Hidden Cost of Prompt Injection Defense
Jun 29, 2026Mitchell Hermon, Rahul Gupta, Weitong Ruan +2LLM Safety Benchmarks Prompt Injection Defense
University of Illinois Urbana-Champaign · 2Amazon
EMPATH: A Multilingual Auditor-Judge Benchmark for Safety Evaluation of Emotional-Support Chatbots
Jun 29, 2026Camilo Chacón SartoriLLM Safety Benchmarks Multilingual Language Model Evaluation
MindSurf
SafePyramid: A Hierarchical Benchmark for In-context Policy Guardrailing
Jun 29, 2026Jiacheng Zhang, Haoyu He, Sen Zhang +5LLM Safety Benchmarks Language Model Safety Evaluation
1ByteDance · 2The University of Melbourne
CAREBench: A Child-Safety Risk Benchmark for Language Models
Jun 29, 2026Kaavya Krishna-Kumar, Elaine Lau, Vaughn Robinson +6LLM Safety Benchmarks Language Model Safety Evaluation
Handshake AI · University of California, Los Angeles · McGill University
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
The Heterogeneous Safety Impacts of Benign Multilingual Fine-Tuning
Jun 27, 2026Will Hawkins, Kaivalya Rawal, Jonathan Rystrøm +8LLM Safety Benchmarks Multilingual Language Model Evaluation
Oxford Internet Institute, University of Oxford · Hasso Plattner Institute, University of Potsdam · Chinese University of Hong Kong +2
Yuvion LLM: An Adversarially-Aware Large Language Model for Content And AI Safety
Jun 26, 2026Ting Ma, Xiufeng Huang, Benlei Cui +43LLM Safety Benchmarks LLM Safety Alignment
Do Encoders Suffice? A Systematic Comparison of Encoder and Decoder Safety Judges for LLM Adversarial Evaluation
Jun 24, 2026Han Jeon, Shiv Medler, Joseph Voyles +1LLM Safety Benchmarks LLM-as-a-Judge
PricewaterhouseCoopers, U.S.A.
FinRED: An Expert-Guided Benchmark Generation and Evaluation Framework for Financial LLM Red-Teaming
Jun 18, 2026Chaeyun Kim, Daeyoung Park, Junghwan Kim +4LLM Safety Benchmarks Financial Services
DATUMO INC., Republic of Korea · Korea Advanced Institute of Science and Technology (KAIST), Republic of Korea · Financial Security Institute (FSI), Republic of Korea
SciRisk-Bench: A Risk-Dimension-Aware Benchmark for AI4Science Safety
Jun 17, 2026Linghao Feng, Yinqian Sun, Dongqi Liang +8LLM Safety Benchmarks AI for Science
Brain-inspired Cognitive Intelligence Lab, Institute of Automation, Chinese Academy of Sciences, Beijing, China · School of Future Technology, University of Chinese Academy of Sciences, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, China +5

Aug 1, 2026, cs.CRJ/K move · Enter open · S save

Single Canonical Prompts Underestimate LLM Safety's Surface-Form Sensitivity

Yongxi Zhou, Junwei Yao, Yuanzhe Liu, Zihan Dong+3

1Northeastern University, Massachusetts, USA · 2Georgia Institute of Technology, Georgia, USA · University of Southern California, California, USA

LLM Safety Benchmarks Prompt Sensitivity Language Model Safety Evaluation Language Model Robustness

Open paper AskPDF arXiv

A benchmark score is a measurement instrument, yet most benchmarks read each item at a single canonical surface form. We ask whether that reading is faithful: when an item's intent is held fixed and only its meaning-preserving surface form varies, does the canonical-form score estimate model behavior well, and how much of any variation is decoding/judge noise rather than signal? We instantiate this in safety, a high-stakes setting with no gold label to average toward. To avoid prior confounds, we pre-author the reformulations (refusal-free, mostly non-LLM: machine back-translation and a Matrix-Language-Frame code-switch generator) so an identical surface form reaches every model, score all responses with one human-anchored, vendor-neutral judge (Claude, kappa = 0.86 vs. human on unsafe compliance, stable across languages, cross-checked by GPT-4o), and verify intent preservation. On 370 seeds x 5 surface forms x 5 models, no single transformation is uniformly most dangerous (6 of 20 per-transformation McNemar tests survive correction, most protective). Yet evaluating only the canonical prompt underestimates unsafe compliance: the union of unsafe outcomes across forms exceeds even the worst single form by 3.3-12.9 pp, with bootstrap 95% CIs excluding zero for all five models, and 5-13% of seeds safe on canonical are unsafe under some reformulation -- above a zero stochasticity floor (canonical resampled five times at temperature 0 gives 0/370 new exposures). The size of this gap is model-dependent (largest on Gemini 2.5 Pro). One form recovers only ~53% of a model's observed unsafe surface and about three reach 85% -- a redundancy characterization of this form set, not of a defined population. A benign control (XSTest) suggests the instability is bidirectional, though the benign and harmful pools are not item-matched. We release the dataset, code, and per-response labels.

Momentum

Related topics

Latest papers 141