SweRank+: Multilingual, Multi-Turn Code Ranking for Software Issue Localization

TL;DR

SweRank+ introduces a multilingual, multi-turn code ranking framework that improves issue localization accuracy in large-scale codebases by combining cross-lingual retrieval, large language model reranking, and iterative reasoning.

Contribution

The paper presents SweRank+, a novel framework integrating cross-lingual code ranking with an agentic, multi-turn search process for better issue localization in multilingual codebases.

Findings

Achieves state-of-the-art results on multilingual issue localization benchmarks.

Multi-turn reasoning with SweRankAgent improves localization accuracy over single-pass methods.

Effective in diverse programming languages, demonstrating broad applicability.

Abstract

Maintaining large-scale, multilingual codebases hinges on accurately localizing issues, which requires mapping natural-language error descriptions to the relevant functions that need to be modified. However, existing ranking approaches are often Python-centric and perform a single-pass search over the codebase. This work introduces SweRank+, a framework that couples SweRankMulti, a cross-lingual code ranking tool, with SweRankAgent, an agentic search setup, for iterative, multi-turn reasoning over the code repository. SweRankMulti comprises a code embedding retriever and a listwise LLM reranker, and is trained using a carefully curated large-scale issue localization dataset spanning multiple popular programming languages. SweRankAgent adopts an agentic search loop that moves beyond single-shot localization with a memory buffer to reason and accumulate relevant localization candidates over multiple turns. Our experiments on issue localization benchmarks spanning various languages demonstrate new state-of-the-art performance with SweRankMulti, while SweRankAgent further improves localization over single-pass ranking.