{"url":"/dataset/bbh","name":"BBH","full_name":"BIG-Bench Hard","description_markdown":"BIG-Bench Hard (BBH) is a subset of the BIG-Bench, a diverse evaluation suite for language models. BBH focuses on a suite of 23 challenging tasks from BIG-Bench that were found to be beyond the capabilities of current language models. These tasks are ones where prior language model evaluations did not outperform the average human-rater.\r\n\r\nThe BBH tasks require multi-step reasoning, and it was found that few-shot prompting without Chain-of-Thought (CoT), as done in the BIG-Bench evaluations, substantially underestimates the best performance and capabilities of language models. When CoT prompting was applied to BBH tasks, it enabled PaLM to surpass the average human-rater performance on 10 of the 23 tasks, and Codex to surpass the average human-rater performance on 17 of the 23 tasks.","description_withheld":null,"homepage":"https://github.com/suzgunmirac/BIG-Bench-Hard","introduced_date":"2022-10-17","introduced_date_note":null,"introduced_by":{"paper":"/paper/challenging-big-bench-tasks-and-whether-chain","title":"Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them","first_author":"Mirac Suzgun","url":null},"license":null,"modalities":[],"tasks":[{"name":"Question Answering","url":"/task/question-answering","datasets_with_task":"/datasets/task/question-answering"},{"name":"task","url":null,"datasets_with_task":"/datasets/task/task"},{"name":"Multi-task Language Understanding","url":"/task/multi-task-language-understanding","datasets_with_task":"/datasets/task/multi-task-language-understanding"}],"languages":[],"variants":["BBH-nlp","BBH-alg","Big-bench Hard","BBH"],"data_loaders":[{"repo":"https://github.com/huggingface/datasets","url":"https://huggingface.co/datasets/weblab-GENIAC/jbbh","frameworks":["tf","pytorch","jax"]},{"repo":"https://github.com/suzgunmirac/big-bench-hard","url":"https://github.com/suzgunmirac/big-bench-hard","frameworks":["tf","pytorch","jax"]}],"num_papers_in_archive":352,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/on-big-bench-hard","task":"","dataset_variant":"Big-bench Hard","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"CoT-T5 11B","paper":"/paper/the-cot-collection-improving-zero-shot-and","metrics":{"Accuracy":"48"},"code_links":[{"title":"kaistai/cot-collection","url":"https://github.com/kaistai/cot-collection"},{"title":"kaist-lklab/cot-collection","url":"https://github.com/kaist-lklab/cot-collection"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"},{"leaderboard":"/sota/question-answering-on-bbh","task":"Question Answering","dataset_variant":"BBH","rows":1,"metrics":["Accuracy"],"first_row_in_archive_order":{"model":"Shakti-LLM (2.5B)","paper":"/paper/shakti-a-2-5-billion-parameter-small-language","metrics":{"Accuracy":"58.2"},"code_links":[]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/shakti-a-2-5-billion-parameter-small-language","title":"SHAKTI: A 2.5 Billion Parameter Small Language Model Optimized for Edge AI and Low-Resource Environments","date":"2024-10-15","rows_on_this_dataset":1,"code_links":0,"syntology":null},{"paper":"/paper/the-cot-collection-improving-zero-shot-and","title":"The CoT Collection: Improving Zero-shot and Few-shot Learning of Language Models via Chain-of-Thought Fine-Tuning","date":"2023-05-23","rows_on_this_dataset":1,"code_links":2,"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."}