Language Model Safety Evaluation

Follow this topic Ask about this topic

Momentum

30 papers in the last four weeks, up 58% on the four weeks before. 0.3% of all new papers.

Jul 13Week of Sep 28

Latest papers 347

All topics

When and How Severely: Scenario-Specific Safety Envelopes for Driving VLAs
Jun 12, 2026Abhinaw Priyadershi, Jelena FrtunikjVLMs for Autonomous Driving Language Model Safety Evaluation
NVIDIA Corporation, Santa Clara, CA, USA · NVIDIA GmbH, Munich, Germany
Right or Wrong, Models Comply: Directional Blindness in LLM Moral Judgment
Jun 12, 2026Jihye Kim, Jeffrey FlaniganLLM Alignment Moral Reasoning in Language Models
University of California, Santa Cruz
RogueAI: A Reverse Turing Test for Detecting Licensed AI Deception in Dialogue
Jun 11, 2026Sara Candussio, Emanuele Ballarin, Lorenzo Bonin +2AI Agent Reliability Language Model Safety Evaluation
1AILab, MIGe, University of Trieste, 34127 Trieste, IT · 2Computational Statistics and Machine Learning, Istituto Italiano di Tecnologia, Genova, IT · 3DIA, University of Trieste, 34127 Trieste, IT
ERTS: Adversarial Robustness Testing of Ethical AI via Semantic Perturbation in a Bounded Consequence Space
Jun 11, 2026Pratyush ChaudhariMoral Reasoning in Language Models Language Model Safety Evaluation
Independent Researcher
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
Schützen: Evaluating LLM Safety in Bulgarian and German Contexts
Jun 9, 2026Kiril Georgiev, Yuxia Wang, Dimitar Iliyanov Dimitrov +2LLM Safety Benchmarks Multilingual Language Model Evaluation
Sofia University "St. Kliment Ohridski" · INSAIT, Sofia University “St. Kliment Ohridski” · MBZUAI
CIAware-Bench: Benchmarking Control Intervention Awareness Across Frontier LLMs
Jun 9, 2026Joachim Schaeffer, Alexander Panfilov, Thomas Jiralerspong +4Language Model Safety Evaluation AI Agent Security Benchmarks
MATS · ELLIS Institute Tübingen, MPI for Intelligent Systems & Tübingen AI Center · Mila – Quebec AI Institute +4
Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models
Jun 9, 2026Prajakta Kini, Avinash Reddy, Souradip Chakraborty +4LLM Alignment Language Model Safety Evaluation
University of Colorado Boulder · University of Central Florida · University of Maryland College Park +1
When the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning Models
Jun 9, 2026Sai Kartheek Reddy Kasu, Nils Lukas, Samuele PoppiCoT Faithfulness LLM Alignment
Independent Researcher · Mohamed bin Zayed University of Artificial Intelligence (MBZUAI)
What the Eyes See, the LLMs Miss: Exploiting Human Perception for Adversarial Text Attacks
Jun 8, 2026Qin Yang, Lu Malloy, Joshua Lee +4Language Model Safety Evaluation Black-Box Adversarial Attacks
University of Connecticut · University of Tennessee · University of California, Santa Barbara +1
Emergent alignment and the projectability of ethical personas
Jun 8, 2026Guillermo Del Pinal, Youngchan Lee, Calum McNamara +1Moral Reasoning in Language Models Language Model Safety Evaluation
University of Massachusetts Amherst · Indiana University Bloomington
Culturally-Adapted Red-Teaming Across East and Southeast Asian Contexts: A Methodological and Comparative Analysis
Jun 8, 2026Hyeji Choi, Yongtaek Lim, Minwoo KimMultilingual Language Model Evaluation Language Model Safety Evaluation
AI Safety Team, DATUMO.INC, Seoul, South Korea.
Diffuse AI Control on Fuzzy Tasks
Jun 8, 2026Mikhail Terekhov, Caglar Gulcehre, Vivek Hebbar +1Adversarial Prompt Generation Language Model Safety Evaluation
Anthropic Fellows Program (via MATS) · EPFL · Redwood Research +1
Activation Steering Induces Emergent Misalignment: A More Comprehensive Evaluation
Jun 7, 2026Qi Cao, Jian Lou, Meiting Liu +4Language Model Safety Evaluation LLM Safety
Nanyang Technological University · Sun Yat-sen University · University of Science and Technology of China +1
Sycophancy as a Multilingual Alignment Failure: How Safety Degrades Across Languages, Topics, and Models
Jun 7, 2026Arya Shah, Himanshu Beniwal, Mayank Singh +1Multilingual Language Models Multilingual Language Model Evaluation
IIT Gandhinagar · Asian Institute of Technology
When Behavioral Safety Evaluation Fails: A Representation-Level Perspective
Jun 6, 2026Enyi Jiang, Anders Gjølbye, Yibo Jacky Zhang +1Language Model Safety Evaluation LLM Auditing
1Stanford University · University of Illinois Urbana-Champaign · 3Technical University of Denmark
Stress-testing medical large language models reveals latent safety pathology beyond benchmark accuracy
Jun 6, 2026Yuan Shen, Xiaojun Wu, Linghua YuLLM Evaluation Language Model Safety Evaluation
College of Computer Science and Technology, Zhejiang University, PR China · The First Hospital of Jiaxing, Zhejiang Province, PR China
Beyond Pass/Fail: Using Process Mining to Understand How LLMs Resist (and Fail) Red Team Attacks
Jun 5, 2026Zvi TopolLanguage Model Safety Evaluation Adversarial Attacks on LLMs
MuyVentive, LLC
CogManip: Benchmarking Manipulative Behavior in Multi-Turn Interactions with Large Language Model
Jun 4, 2026Zeyang Yue, Chenfei Yan, Feifei Zhao +5Language Model Safety Evaluation LLM Auditing
School of Artificial Intelligence, Beihang University · 2BrainCog AI Lab, CASIA · School of Artificial Intelligence, UCAS +4
When Should Memory Stay Silent: Measuring Memory-Use Boundaries in Memory-Augmented Conversational Agents
Jun 4, 2026Lingxiang Xu, Jiaoyun Yang, Min Hu +2Language Model Safety Evaluation Persistent Memory for Language Models
†Hefei University of Technology · ‡Harvard Medical School
SpeechJBB: Probing Safety Alignment and Comprehension in Large Audio Language Models under Code-Switched Speech
Jun 4, 2026Virginia Ceccatelli, Yejin Jeon, David Ifeoluwa AdelaniAudio-Language Model Evaluation Language Model Safety Evaluation
Mila - Quebec AI Institute · McGill University, Canada · Canada CIFAR AI Chair
Beyond WER: A Paired Acoustic Stress Test for Ambient Clinical Scribes
Jun 4, 2026Xiao-Hang Jiang, Han-Jie Guo, Ying-Si Liang +4ASR Evaluation Language Model Safety Evaluation
University of Science and Technology of China, China · iFLYTEK Co., Ltd., China
Voices Across Registers: Corpus-Conditioned Vernacular Jailbreaks against Aligned LLMs via Fanfiction Subgenres
Jun 3, 2026Zhongze Luo, Ruihe Shi, Zhenshuai Yin +3Adversarial Prompt Generation Language Model Safety Evaluation
School of Science and Engineering, The Chinese University of Hong Kong (Shenzhen) · School of Microelectronics, Xi’an Jiaotong University · 2The Shenzhen Future Network of Intelligence Institute (FNii-Shenzhen) +1
Safety Measurements for Fine-tuned LLMs Should be Grounded in Capability
Jun 2, 2026Krishnapriya Vishnubhotla, Hillary Dawkins, Isar Nejadgholi +1Language Model Safety Evaluation LLM Fine-Tuning
National Research Council, Canada
Evaluating LLMs' Effectiveness on Real-World Consumer Device Repair Questions
Jun 2, 2026Atm Mizanur Rahman, Md Arid Hasan, Syed Ishtiaque Ahmed +1Multilingual Language Model Evaluation LLM Evaluation
University of Illinois Urbana-Champaign · University of Toronto
TriEval: A Resource-Efficient Pipeline for LLM Bias, Toxicity, and Truthfulness Assessment
Jun 2, 2026Akshatha Srikantha, Manpreet Singh, Yash Jajoo +1LLM Evaluation Language Model Safety Evaluation
University of California, Irvine, CA, USA · Boston University, Boston, MA, USA · New York University, New York City, NY, USA +1
Gate AI: LLM Security Benchmark Evaluation Methodology and Results
Jun 1, 2026Ryle Goehausen, Marcus SousaLLM Safety Benchmarks Language Model Safety Evaluation
Acceptance-Test-Driven Evaluation Protocols for Business-Centric LLM Systems
Jun 1, 2026Eric LiangLLM Evaluation Language Model Safety Evaluation
Oracle
Investigating and Alleviating Harm Amplification in LLM Interactions
Jun 1, 2026Ruohao Guo, Wei Xu, Alan RitterLanguage Model Safety Evaluation LLM Safety
Georgia Institute of Technology
SentGuard: Sentence-Level Streaming Guardrails for Large Language Models
Jun 1, 2026Jiaqi Yu, Xin Wang, Yixu Wang +4Language Model Safety Evaluation LLM Guardrails
1Fudan University · 2Shanghai AI Laboratory

Jun 12, 2026, cs.ROJ/K move · Enter open · S save

When and How Severely: Scenario-Specific Safety Envelopes for Driving VLAs

Abhinaw Priyadershi, Jelena Frtunikj

NVIDIA Corporation, Santa Clara, CA, USA · NVIDIA GmbH, Munich, Germany

VLMs for Autonomous Driving Language Model Safety Evaluation Autonomous Driving Safety Evaluation Robot Safety AI Safety Evaluation

Open paper AskPDF arXiv

Safety certification of Vision-Language-Action (VLA) driving planners under ISO 21448 (SOTIF) rests on an Operational Design Domain (ODD) specification that answers two complementary questions: when does the planner start to fail, and how severely does it fail once it does? We evaluate Alpamayo R1, a 10B-parameter open-weight driving VLA, on 15,968 (clip, attack) pairs. We find a conservative-aggregate gap: an aggregate safe threshold of

σ\leq 50

under a 15% average displacement error (ADE) budget masks well-sampled scenarios that tolerate the top of the tested grid (

σ= 70

). A Gaussian Mixture Model (GMM) on the changed-explanation subset identifies six discrete severity bands (BIC-optimal

k{=}6

), so two perturbation conditions with the same mean error can differ materially in their share of high-severity (C4/C5) failures. Joining the two analyses on the same corpus surfaces a finding neither yields in isolation: the scenarios with the loosest noise thresholds are not those with the lowest high-severity rate: STOP_SIGNAL concentrates roughly

4\times

the C4/C5 share of LANE_KEEPING despite tolerating a larger

σ

. A deployable SOTIF ODD specification for driving VLAs therefore requires a two-dimensional safety envelope, not a single aggregate value per hazard.

Momentum

Related topics

Latest papers 347