LARGER: Lexically Anchored Repository Graph Exploration and Retrieval

TL;DR

LARGER is a framework that enhances repository navigation for coding agents by converting lexical matches into structural context, improving localization and understanding without external graph tools.

Contribution

It introduces a lexically anchored active-set retrieval method that integrates structural graph exploration into existing search loops for better code repository localization.

Findings

Improves file-level Acc@5 on LocBench by +13.9 points with tuned hyperparameters.

Gains +11.8 points on LocBench with fixed hyperparameters over the strongest baseline.

Delivers consistent improvements across multiple benchmarks including MuLocBench and SWE-Atlas.

Abstract

Repository-level coding agents must first localize the files and symbols relevant to a task; failures at this stage can cascade across downstream objectives ranging from patch generation to test writing and codebase question answering. Existing agents navigate repositories primarily through lexical search, often missing structural relations such as imports, call chains, type hierarchies, and code-test links. Graph-based retrieval can recover such dependencies, but existing approaches often require separate graph tools or traversal stages that fragment the agent's interaction loop. We formalize repository context localization as Lexically Anchored Structural Localization, where success depends on turning lexical matches into high-precision structural entry points and exposing the most useful confidence-filtered local neighborhoods within the agent's existing search loop. We introduce LARGER (Lexically Anchored Repository Graph Exploration and Retrieval), a lexically anchored active-set retrieval framework that starts from lexical matches, aligns them to graph anchors, and performs confidence-filtered local expansion within the agent's existing search loop. LARGER integrates directly into existing CLI coding agents without requiring external graph databases or specialized graph interfaces. Across four benchmarks spanning localization, test generation, and codebase understanding, LARGER improves file-level Acc@5 on LocBench by +13.9 points with tuned hyperparameters and still gains +11.8 points with fixed hyperparameters over the strongest baseline, while delivering consistent gains on MuLocBench, SWE-Atlas Test Writing, and SWE-Atlas Codebase QA.