Researchers have developed Chronos, a new framework designed to help large language model (LLM)-based code agents reason about software evolution. Chronos distills historical pull requests into structured 'experience cards' and connects them via a graph of code, developer intent, and organizational relations. This allows semantic search to identify relevant changes, which are then retrieved through multi-hop expansion for selective reading. The framework guides agents in generating and selecting patches, showing significant improvements on benchmarks like SWE-Bench Verified, SWE-Bench Pro, and FEA-Bench Lite. AI
IMPACT Enhances LLM code agents' ability to understand and utilize historical code changes, potentially improving automated software development and maintenance.
RANK_REASON The cluster contains a research paper detailing a new framework for LLM code agents. [lever_c_demoted from research: ic=1 ai=1.0]
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →