Artifact-to-Capability: Diagnosing Breaks in Capability Formation of Self-Evolving Agents in a Controlled Environment
cs.AI
Submitted: 2026-11-03
Updated: 2026-11-03
Terminology
Sources
- VRL-Bench: Benchmarking agents on computer control tasks under finite trial budgets
- Agent Skills Can Be Harmful: An Empirical Study of Skill-Induced Failures in LLM Agents
- EvoAgentBench: Benchmarking Agent Self-Evolution via Ability Transfer
- Skill Use or Skill Theater? Evaluating the Reasoning Backroom in Skill-Augmented Language Agents
- BenchTrace: A Benchmark for Testing Reflection Ability and Controlled Evolution in LLM Agents
- SEA-Eval: A Benchmark for Evaluating Self-Evolving Agents Beyond Episodic Assessment
- How Much Heavy Lifting Can an Agent Harness Do?: Measuring the LLM's Residual Role in a Planning Agent
- Beyond Endpoint Performance: Process-Level Evaluation of Self-Evolving Agents
- Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
- What Do Evolutionary Coding Agents Evolve?
- The Why Behind the Action: Unveiling Internal Drivers via Agentic Attribution
- When Self-Evolution Backfires: Pre-Commit Gating against Skill Contamination in LLM Agents
- Agent Plasticity: Measuring Self-Improvement Through Experience
- SAGE: A Statistical Acceptance Gate for Self-Evolving Agents
- Not All Skills Help: Measuring and Repairing Agent Knowledge
- The Rise and Potential of Large Language Model Based Agents: A Survey
- How Memory Management Impacts LLM Agents: An Empirical Study of Experience-Following Behavior
- One Skill Too Many: How Co-Installed Skills Conflict in Coding Agents
- Useful Memories Become Faulty When Continuously Updated by LLMs
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