DocsChisel adaptively adds, removes, and refines information fields in tool documentation based on failed execution traces, improving LLM agent task success rate by 95.89% over original documentation and 75.15% over prior methods.
(2025) Claude code
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
DOCSCHISEL: Adaptive Tool Documentation Optimization Framework for LLM Agents
DocsChisel adaptively adds, removes, and refines information fields in tool documentation based on failed execution traces, improving LLM agent task success rate by 95.89% over original documentation and 75.15% over prior methods.