Recall Before Rerank: Benchmarking Deep Learning Models for Large-Scale Code-to-Code Retrieval

2026-06-24

AI Summary

This work addresses the precision and scalability bottlenecks faced by current deep learning models in the initial recall stage over terabyte-scale code corpora, which hinder efficient semantic search and clone detection. For the first time, we systematically evaluate the recall capabilities of large code models on a massive multilingual code dataset and propose a lightweight LLM-driven approach that leverages code normalization and query rewriting to enhance retrieval effectiveness. Experimental results demonstrate that our method substantially improves recall accuracy for lower-capacity models, while also uncovering critical limitations of existing approaches in cross-dataset generalization and deployment under resource constraints. These findings offer practical guidance for building efficient and scalable code retrieval systems.

Abstract

Semantic code search and clone detection are essential for software development, maintenance, and reuse. This paper evaluates the effectiveness, efficiency, and scalability of contemporary deep learning models for first-stage recall in large-scale code-to-code search engines. Benchmarking across multiple programming languages and datasets reveals critical limits in the precision and scalability of these models on Terabyte-scale source-code collections. We present LLM-based code normalisation and query-rewriting schemes that yield significant gains in precision for lower-performing models. Our results question the sustainability of resource-constrained deployment and the assumed robustness of current code-specialised LLMs across datasets. We conclude with actionable insights for building scalable, efficient code-retrieval systems.

Problem

• code-to-code retrieval
• deep learning models
• large-scale benchmarking
• recall
• scalability

Innovation

• code-to-code retrieval
• first-stage recall
• LLM-based normalization
• query rewriting
• scalable code search