CoREB is a contamination-limited multitask benchmark for code search covering text-to-code, code-to-text and code-to-code tasks, with a fine-tuned reranker that delivers consistent gains where prior models do not.
Because CodeSearchNet has served as public training data since 2019, models evaluated on it face severe contamination risk (Allamanis, 2019; Hernandez Lopez et al., 2024)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Beyond Retrieval: A Multitask Benchmark and Model for Code Search
CoREB is a contamination-limited multitask benchmark for code search covering text-to-code, code-to-text and code-to-code tasks, with a fine-tuned reranker that delivers consistent gains where prior models do not.