Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Xue, Siqiao, Liao, Zihan, Qin, Jin, Zhang, Ziyin, Mu, Yixiang, Zhou, Fan, Yu, Hang
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2605.04615
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914542089404416
author Xue, Siqiao
Liao, Zihan
Qin, Jin
Zhang, Ziyin
Mu, Yixiang
Zhou, Fan
Yu, Hang
author_facet Xue, Siqiao
Liao, Zihan
Qin, Jin
Zhang, Ziyin
Mu, Yixiang
Zhou, Fan
Yu, Hang
contents Code search has usually been evaluated as first-stage retrieval, even though production systems rely on broader pipelines with reranking and developer-style queries. Existing benchmarks also suffer from data contamination, label noise, and degenerate binary relevance. In this paper, we introduce \textsc{CoREB}, a contamination-limited, multitask \underline{co}de \underline{r}etrieval and r\underline{e}ranking \underline{b}enchmark, together with a fine-tuned code reranker, that goes beyond retrieval to cover the full code search pipeline. \textsc{CoREB} is built from counterfactually rewritten LiveCodeBench problems in five programming languages and delivered as timed releases with graded relevance judgments. We benchmark eleven embedding models and five rerankers across three tasks: text-to-code, code-to-text, and code-to-code. Our experiments reveal that: \circone code-specialised embeddings dominate code-to-code retrieval (${\sim}2{\times}$ over general encoders), yet no single model wins all three tasks; \circtwo short keyword queries, the format closest to real developer search, collapse every model to near-zero nDCG@10; \circthree off-the-shelf rerankers are task-asymmetric, with a 12-point swing on code-to-code and no baseline net-positive across all tasks; \circfour our fine-tuned \textsc{CoREB-Reranker} is the first to achieve consistent gains across all three tasks. The data and model are released.
format Preprint
id arxiv_https___arxiv_org_abs_2605_04615
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Beyond Retrieval: A Multitask Benchmark and Model for Code Search
Xue, Siqiao
Liao, Zihan
Qin, Jin
Zhang, Ziyin
Mu, Yixiang
Zhou, Fan
Yu, Hang
Software Engineering
Artificial Intelligence
Code search has usually been evaluated as first-stage retrieval, even though production systems rely on broader pipelines with reranking and developer-style queries. Existing benchmarks also suffer from data contamination, label noise, and degenerate binary relevance. In this paper, we introduce \textsc{CoREB}, a contamination-limited, multitask \underline{co}de \underline{r}etrieval and r\underline{e}ranking \underline{b}enchmark, together with a fine-tuned code reranker, that goes beyond retrieval to cover the full code search pipeline. \textsc{CoREB} is built from counterfactually rewritten LiveCodeBench problems in five programming languages and delivered as timed releases with graded relevance judgments. We benchmark eleven embedding models and five rerankers across three tasks: text-to-code, code-to-text, and code-to-code. Our experiments reveal that: \circone code-specialised embeddings dominate code-to-code retrieval (${\sim}2{\times}$ over general encoders), yet no single model wins all three tasks; \circtwo short keyword queries, the format closest to real developer search, collapse every model to near-zero nDCG@10; \circthree off-the-shelf rerankers are task-asymmetric, with a 12-point swing on code-to-code and no baseline net-positive across all tasks; \circfour our fine-tuned \textsc{CoREB-Reranker} is the first to achieve consistent gains across all three tasks. The data and model are released.
title Beyond Retrieval: A Multitask Benchmark and Model for Code Search
topic Software Engineering
Artificial Intelligence
url https://arxiv.org/abs/2605.04615