When Do Summaries Become Instructions? Unsupported Policy in LLM Agents' Self-Written Handoff Summaries
cs.CL
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/Chuw1/handoff-audit
Terminology
Sources
- Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
- Agent Memory Is a Surface for Endogenous Authorization Laundering
- Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents
- Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents
- When Evidence Changes: Evaluating Memory Repair and Re-reading in Language-Model Agents
- Memory-Induced Tool-Drift in LLM Agents
- Kimi K2.5: Visual Agentic Intelligence
- Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
- Safe Multi-Agent Behavior Must Be Maintained, Not Merely Asserted: Constraint Drift in LLM-Based Multi-Agent Systems
- gpt-oss-120b & gpt-oss-20b Model Card
- Red-Teaming Auto Mode: Improving Blocking Classifiers Against Malign Coding Agents
- Is Your Agent Playing Dead? Deployed LLM Agents Exhibit Constraint-Evasive Fabrication and Thanatosis
- Qwen3 Technical Report
- When Memory Becomes Authority: Benchmarking Authority Collapse at the Memory Consolidation Boundary
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data