StateWise: Diagnosing and Repairing Persistent Operational State Before Agent Actions
cs.SE, cs.AI
Submitted: 2026-10-04
Updated: 2026-10-04
Comments: 21 pages, 8 figures
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Remembering More, Risking More: Longitudinal Safety Risks in Memory-Equipped LLM Agents
- SWE-chat: Coding Agent Interactions From Real Users in the Wild
- Ghost in the Agent: Redefining Information Flow Tracking for LLM Agents
- STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?
- SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing
- Ask or Assume? Uncertainty-Aware Clarification-Seeking in Coding Agents
- AgentCanary: A Security Evaluation Framework for Autonomous AI Agents in Real Executable Environments
- MemLineage: Lineage-Guided Enforcement for LLM Agent Memory
- MemGPT: Towards LLMs as Operating Systems
- Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents
- The Why Behind the Action: Unveiling Internal Drivers via Agentic Attribution
- Safeguarding LLM Agents from Misalignment through Provenance Analysis
- Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge
Related papers
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
- PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
- IntentCoding: Amplifying User Intent in Code Generation
- Incentives and Outcomes in Bug Bounties