An LLM-based system translates English code search questions into Semgrep and GQL queries, reaching 55-70% precision/recall on a new synthetic 400-query Java benchmark.
Neural Code Search Evaluation Dataset
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
There has been an increase of interest in code search using natural language. Assessing the performance of such code search models can be difficult without a readily available evaluation suite. In this paper, we present an evaluation dataset consisting of natural language query and code snippet pairs, with the hope that future work in this area can use this dataset as a common benchmark. We also provide the results of two code search models ([1] and [6]) from recent work. The evaluation dataset is available at https://github.com/facebookresearch/Neural-Code-Search-Evaluation-Dataset
citation-role summary
dataset 1
citation-polarity summary
fields
cs.SE 1years
2025 1verdicts
CONDITIONAL 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
Structural Code Search using Natural Language Queries
An LLM-based system translates English code search questions into Semgrep and GQL queries, reaching 55-70% precision/recall on a new synthetic 400-query Java benchmark.