Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Chaudhari, Shravan, Jacob, Rahul Thomas, Goswami, Mononito, Cao, Jiajun, Rashid, Shihab, Bock, Christian
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2512.16956
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917252035510272
author Chaudhari, Shravan
Jacob, Rahul Thomas
Goswami, Mononito
Cao, Jiajun
Rashid, Shihab
Bock, Christian
author_facet Chaudhari, Shravan
Jacob, Rahul Thomas
Goswami, Mononito
Cao, Jiajun
Rashid, Shihab
Bock, Christian
contents Retrieving code functions, classes or files that are relevant in order to solve a given user query, bug report or feature request from large codebases is a fundamental challenge for Large Language Model (LLM)-based coding agents. Agentic approaches typically employ sparse retrieval methods like BM25 or dense embedding strategies to identify semantically relevant units. While embedding-based approaches can outperform BM25 by large margins, they often don't take into consideration the underlying graph-structured characteristics of the codebase. To address this, we propose SpIDER (Spatially Informed Dense Embedding Retrieval), an enhanced dense retrieval approach that integrates LLM-based reasoning along with auxiliary information obtained from graph-based exploration of the codebase. We further introduce SpIDER-Bench, a graph-structured evaluation benchmark curated from SWE-PolyBench, SWEBench-Verified and Multi-SWEBench, spanning codebases from Python, Java, JavaScript and TypeScript programming languages. Empirical results show that SpIDER consistently improves dense retrieval performance by at least 13% across programming languages and benchmarks in SpIDER-Bench.
format Preprint
id arxiv_https___arxiv_org_abs_2512_16956
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SpIDER: Spatially Informed Dense Embedding Retrieval for Software Issue Localization
Chaudhari, Shravan
Jacob, Rahul Thomas
Goswami, Mononito
Cao, Jiajun
Rashid, Shihab
Bock, Christian
Software Engineering
Machine Learning
Retrieving code functions, classes or files that are relevant in order to solve a given user query, bug report or feature request from large codebases is a fundamental challenge for Large Language Model (LLM)-based coding agents. Agentic approaches typically employ sparse retrieval methods like BM25 or dense embedding strategies to identify semantically relevant units. While embedding-based approaches can outperform BM25 by large margins, they often don't take into consideration the underlying graph-structured characteristics of the codebase. To address this, we propose SpIDER (Spatially Informed Dense Embedding Retrieval), an enhanced dense retrieval approach that integrates LLM-based reasoning along with auxiliary information obtained from graph-based exploration of the codebase. We further introduce SpIDER-Bench, a graph-structured evaluation benchmark curated from SWE-PolyBench, SWEBench-Verified and Multi-SWEBench, spanning codebases from Python, Java, JavaScript and TypeScript programming languages. Empirical results show that SpIDER consistently improves dense retrieval performance by at least 13% across programming languages and benchmarks in SpIDER-Bench.
title SpIDER: Spatially Informed Dense Embedding Retrieval for Software Issue Localization
topic Software Engineering
Machine Learning
url https://arxiv.org/abs/2512.16956