LLM Safety Monitoring

LLM: Large Language Model

Follow this topic Ask about this topic

Latest papers

38 papers5 papers in the last four weeks, level with the four weeks before. 0.0% of all new papers.

Classifier Context Rot: Monitor Performance Degrades with Context Length
May 12, 2026Sam Martin, Fabien RogerLong-Context Modeling Long-Context Language Model Evaluation
Anthropic Fellows Program · Anthropic
GLiGuard: Schema-Conditioned Classification for LLM Safeguard
May 8, 2026Urchade Zaratiana, Mary Newhauser, George Hurn-Maloney +1Jailbreak Defense Text Classification
Fastino Labs
ProbeLogits: Kernel-Level LLM Inference Primitives for AI-Native Operating Systems
Apr 13, 2026Daeyeon SonAI Safety LLM Guardrails
Independent Researcher Republic of Korea
Predict, Don't React: Value-Based Safety Forecasting for LLM Streaming
Apr 5, 2026Pride Kavumba, Koki Wataoka, Huy H. Nguyen +2LLM Guardrails LLM Safety Monitoring
SB Intuitions Corp.
SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do
Mar 28, 2026Aditya Dhodapkar, Farhaan PishoriAI Agent Safety LLM Agent Safety
Rensselaer Polytechnic Institute Department of Computer Science · Santa Clara University Department of Engineering
Beyond Content Safety: Real-Time Monitoring for Reasoning Vulnerabilities in Large Language Models
Mar 26, 2026Xunguang Wang, Yuguang Zhou, Qingyue Wang +5LLM Safety Monitoring LLM Safety
The Hong Kong University of Science and Technology, Hong Kong, China · Zhejiang University of Technology, Hangzhou, Zhejiang Province, China
ThaiSafetyBench: Assessing Language Model Safety in Thai Cultural Contexts
Mar 5, 2026Trapoom Ukarapol, Nut Chukamphaeng, Kunat Pipatanakul +1Language Model Safety Evaluation LLM Safety
SCB DataX · Department of Computer Science and Technology, Tsinghua University · SCBX R&D +1
The Ends Justify the Thoughts: RL-Induced Motivated Reasoning Fools CoT Monitors
Oct 20, 2025Nikolaus Howe, Micah CarrollLLM Safety Monitoring CoT Monitoring
Mila, Université de Montréal · UC Berkeley

May 12, 2026J/K move · Enter open · S save

Classifier Context Rot: Monitor Performance Degrades with Context Length

Sam Martin, Fabien Roger

Anthropic Fellows Program · Anthropic

Long-Context Modeling Long-Context Language Model Evaluation AI Agent Monitoring Language Model Safety Evaluation Classification LLM Reliability+1

Open paper AskPDF arXiv

Monitoring coding agents for dangerous behavior using language models requires classifying transcripts that often exceed 500K tokens, but prior agent monitoring benchmarks rarely contain transcripts longer than 100K tokens. We show that when used as classifiers, current frontier models fail to notice dangerous actions more often in longer transcripts. In particular, on a dataset that requires identifying when a coding agent takes a subtly dangerous action, Opus 4.6, GPT 5.4, and Gemini 3.1 miss these actions

2\times

30\times

more often when they occur after 800K tokens of benign activity than when they occur on their own. We also show that these weaknesses can be partially mitigated with prompting techniques such as periodic reminders throughout the transcript and may be mitigated further with better post-training. Monitor evaluations that do not consider long-context degradation are likely overestimating monitor performance.