{"url":"/dataset/longbench","name":"LongBench","full_name":null,"description_markdown":"Click to add a brief description of the dataset (Markdown and LaTeX enabled).\r\n\r\nProvide:\r\n\r\n* a high-level explanation of the dataset characteristics\r\n* explain motivations and summary of its content\r\n* potential use cases of the dataset","description_withheld":null,"homepage":"https://github.com/THUDM/LongBench","introduced_date":null,"introduced_date_note":null,"introduced_by":null,"license":null,"modalities":[],"tasks":[{"name":"Long-Context Understanding","url":"/task/long-context-understanding","datasets_with_task":"/datasets/task/long-context-understanding"}],"languages":[],"variants":["LongBench"],"data_loaders":[],"num_papers_in_archive":8,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/long-context-understanding-on-longbench","task":"Long-Context Understanding","dataset_variant":"LongBench","rows":3,"metrics":["Average Score"],"first_row_in_archive_order":{"model":"GALI(Llama3-8b-ins-4k-to-16k)","paper":"/paper/a-training-free-length-extrapolation-approach","metrics":{"Average Score":"46.22"},"code_links":[{"title":"academycityl/gali","url":"https://github.com/academycityl/gali"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/a-training-free-length-extrapolation-approach","title":"A Training-Free Length Extrapolation Approach for LLMs: Greedy Attention Logit Interpolation (GALI)","date":"2025-02-04","rows_on_this_dataset":3,"code_links":1,"syntology":null}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":0,"samples_harvested":0,"samples_ran":0,"samples_unverified":0,"pointer_only_for_licence":0,"papers_with_no_sample_that_ran":0,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}