Beyond Skill Evolution: Self-Evolving Context Management Policies for Long-Horizon Agent Harnesses
cs.AI
Submitted: 2026-09-28
Updated: 2026-09-28
Sources
- EvoSkill: Automated Skill Discovery for Multi-Agent Systems
- AgentRx: Diagnosing AI Agent Failures from Execution Trajectories
- Reinforcement Learning for Long-Horizon Interactive LLM Agents
- When Continual Learning Moves to Memory: A Study of Experience Reuse in LLM Agents
- DeepSWE: Measuring Frontier Coding Agents on Original, Long-Horizon Engineering Tasks
- HarnessEvolve: Learning from Reference Trajectories for Reliable Agent Self-Evolution
- ACON: Optimizing Context Compression for Long-horizon LLM Agents
- Meta-Harness: End-to-End Optimization of Model Harnesses
- Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents
- ACM: Agentic Context Management for Long Horizon Tasks
- Long-Horizon-Terminal-Bench: Testing the Limits of Agents on Long-Horizon Terminal Tasks with Dense Reward-Based Grading
- Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
- LongHorizon-Harness: Advancing Long-Horizon Agents for Real-World Tasks
- Towards Reliable Context Compression for Long-Horizon Agents: An Empirical Study of Execution Instability
- Retrieval-Augmented LLM Agents: Learning to Learn from Experience
- TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
- A Self-Evolving Framework for Efficient Terminal Agents via Observational Context Compression
- Causal Agent Replay: Counterfactual Attribution for LLM-Agent Failures
- Evo-Harness: Context-to-Harness Skill Compilation for Self-Evolving Agents
Related papers
- 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
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems