Research
Retrieval-Oriented Code Representations Sharpen Agentic Bug Localization
LLM coding agents live or die on retrieving the right code context, yet how textual code representations affect retrieval and localization has been underexplored. This paper reframes file-level bug localization as a representation-driven retrieval problem and studies which code representations best surface the right files for repository-level tasks. The practical implication: how you encode and index code for retrieval may matter as much as the agent's reasoning when localizing bugs at repo scale.
Source
↳ Follow the thread