Cleaning an Agent's History Hurts When the World Still Holds the Error
cs.AI, cs.CL, cs.LG
Submitted: 2026-11-10
Updated: 2026-11-10
Terminology
Sources
- From Confident Closing to Silent Failure: Characterizing False Success in LLM Agents
- The Self-Correction Illusion: Role Relabeling Gates Explicit Error Flagging in Large Language Models
- AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
- Do Not Restart: Residual Completion for Stateful Agent Handoffs
- The Replay Gap: Static Evaluation of Model Switching in LLM Agents Scores the Wrong World
- Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection
- Self-Preference Is Weak or Absent in Verifiable Instruction-Following Revision: A Four-Model Test Under Genuine Authorship
- Feedback That Backfires: Why Small Language Model Agents Repeat the Call They Just Watched Fail
- The Memory Trust Gap: Reliance on Stale Evidence Across Model Scales in LLM Agents
- Large Language Models Cannot Self-Correct Reasoning Yet
- Does Fault Localization Beat a Fresh Attempt? A Placebo-Controlled Study of Test-Guided Code Repair
- ACON: Optimizing Context Compression for Long-horizon LLM Agents
- Handoff Debt: The Rediscovery Cost When Coding Agents Take Over Interrupted Tasks
- Generator-Assistant Stepwise Rollback Framework for Large Language Model Agent
- REFLECT: Intervention-Supported Error Attribution for Silent Failures in LLM Agent Traces
- Lost in the Middle: How Language Models Use Long Contexts
- AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories
- Atomix: Timely, Transactional Tool Use for Reliable Agentic Workflows
- LLM Evaluators Recognize and Favor Their Own Generations
- Quantifying Label-Induced Bias in Large Language Model Self- and Cross-Evaluations
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents