FEA-Bench

Dataset Card for FEA-Bench A Benchmark for Evaluating Repository-Level Code Generation for Feature Implementation. Dataset Details Dataset Description The FEA-Bench is a benchmark with a test set that contains 1,401 task instances from 83 Github repositories. This benchmark aims to evaluate the capabilities of reposito

Papers using FEA-Bench (2)

FEA-Bench: A Benchmark for Evaluating Repository-Level Code Generation for Feature Implementation2025 · 3 citesFEA-Bench: A Benchmark for Evaluating Repository-Level Code Generation for Feature Implementation2025 · 3 cites