Trace2Skill evolves task-specific skills from rollout traces and verifier feedback to boost hardware LLM agent performance on hard CVDP tasks without model fine-tuning.
AgentBench: Evaluating LLMs as agents
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Trace2Skill: Verifier-Guided Skill Evolution for Long-Context EDA Agents
Trace2Skill evolves task-specific skills from rollout traces and verifier feedback to boost hardware LLM agent performance on hard CVDP tasks without model fine-tuning.