xiaoy1028/SWE-bench_Lite_bm25_500k
SWE-bench_Lite Generated via https://github.com/SWE-bench/SWE-bench/blob/main/swebench/inference/make_datasets/README.md Set k=100 to generated the bm25 search results first, and then use both k=100 and max_context_len=500000 to generate the dataset. Patch file recall: 90.67; higher than the official princeton-nlp/SWE-bench_Lite_bm25_13K (recall=33.67) and princeton-nlp/SWE-bench_Lite_bm25_27K (recall=49.00)
SWE-benchLite Generated via https://github.com/SWE-bench/SWE-bench/blob/main/swebench/inference/makedatasets/README.md
Set k=100 to generated the bm25 search results first, and then use both k=100 and maxcontextlen=500000 to generate the dataset.
Patch file recall: 90.67; higher than the official princeton-nlp/SWE-benchLitebm2513K (recall=33.67) and princeton-nlp/SWE-benchLitebm2527K (recall=49.00)