{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/musique-multi-hop-questions-via-single-hop","title":"MuSiQue: Multihop Questions via Single-hop Question Composition","arxiv_id":"2108.00573","date":"2021-08-02","proceeding":null,"authors":["Harsh Trivedi","Niranjan Balasubramanian","Tushar Khot","Ashish Sabharwal"],"abstract":"Multihop reasoning remains an elusive goal as existing multihop benchmarks are known to be largely solvable via shortcuts. Can we create a question answering (QA) dataset that, by construction, \\emph{requires} proper multihop reasoning? To this end, we introduce a bottom-up approach that systematically selects composable pairs of single-hop questions that are connected, i.e., where one reasoning step critically relies on information from another. This bottom-up methodology lets us explore a vast space of questions and add stringent filters as well as other mechanisms targeting connected reasoning. It provides fine-grained control over the construction process and the properties of the resulting $k$-hop questions. We use this methodology to create MuSiQue-Ans, a new multihop QA dataset with 25K 2-4 hop questions. Relative to existing datasets, MuSiQue-Ans is more difficult overall (3x increase in human-machine gap), and harder to cheat via disconnected reasoning (e.g., a single-hop model has a 30 point drop in F1). We further add unanswerable contrast questions to produce a more stringent dataset, MuSiQue-Full. We hope our datasets will help the NLP community develop models that perform genuine multihop reasoning.","url_abs":"https://arxiv.org/abs/2108.00573v3","url_pdf":"https://arxiv.org/pdf/2108.00573v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"musique-multi-hop-questions-via-single-hop","repo_url":"https://github.com/stonybrooknlp/musique","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"musique-multi-hop-questions-via-single-hop","repo_url":"https://github.com/Bodoral/MuSiQue_replication","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"paper_slug":"musique-multi-hop-questions-via-single-hop","repo_url":"https://github.com/mangopy/direct-rag-learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"multi-hop-question-answering","task_name":"Multi-hop Question Answering"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[{"slug":"musique-ans","name":"MuSiQue-Ans","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2108.00573","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2108.00573"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/mangopy/direct-rag-learning","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Bodoral/MuSiQue_replication","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/stonybrooknlp/musique","reach":null}],"summary":{"ran_draft_wrong":1,"unverified":15},"by_repo_kind":{"official":{"samples":1,"ran":1,"repositories":1},"listed":{"samples":15,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"a0c3c47032ca66fd","entry":"read_jsonl","repo":"stonybrooknlp/musique","repo_kind":"official","path":"evaluate_v1.0.py","file_url":"https://github.com/stonybrooknlp/musique/blob/HEAD/evaluate_v1.0.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"CC-BY-4.0","inline_ok":false,"mcp_get_code":{"code_sha256":"a0c3c47032ca66fd"}},{"code_sha256_prefix":"ebfd9b0ec5f4b3ef","entry":"cal_ppl","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/entropy.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/entropy.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ebfd9b0ec5f4b3ef"}},{"code_sha256_prefix":"fac0d9d6ff56e2b2","entry":"causal_cross_entropy","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/entropy.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/entropy.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fac0d9d6ff56e2b2"}},{"code_sha256_prefix":"9369e71b95149aa5","entry":"collect_docs","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/evaluate.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/evaluate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9369e71b95149aa5"}},{"code_sha256_prefix":"4a62a7015c03e433","entry":"document_retrieval","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/prepare.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/prepare.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4a62a7015c03e433"}},{"code_sha256_prefix":"0b2979a7d1760cb1","entry":"eval_em","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/utilize/metrics.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/utilize/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0b2979a7d1760cb1"}},{"code_sha256_prefix":"4e8dbe891b051948","entry":"gather_data","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/prepare.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/prepare.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4e8dbe891b051948"}},{"code_sha256_prefix":"bec7c35aef9d72d5","entry":"generation_instruction","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/instruction.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/instruction.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bec7c35aef9d72d5"}},{"code_sha256_prefix":"a54293fda5579b8c","entry":"get_command","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/sampling.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/sampling.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"a54293fda5579b8c"}},{"code_sha256_prefix":"bb3452c00b70772e","entry":"launch","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/evaluate.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/evaluate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bb3452c00b70772e"}},{"code_sha256_prefix":"97cb57dd2386abcb","entry":"normalize_answer","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/utilize/metrics.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/utilize/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"97cb57dd2386abcb"}},{"code_sha256_prefix":"ae22245c5794b39a","entry":"prepare_warmup_data_for_generation","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/prepare.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/prepare.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"ae22245c5794b39a"}},{"code_sha256_prefix":"9b674a8aeb03a2f2","entry":"preprocess_supervised_dataset","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/prepare/entropy.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/prepare/entropy.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9b674a8aeb03a2f2"}},{"code_sha256_prefix":"203ee66a16f129c1","entry":"rank_instruction","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/instruction.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/instruction.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"203ee66a16f129c1"}},{"code_sha256_prefix":"1ac4537f9c4bf017","entry":"rounder","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/utilize/metrics.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/utilize/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1ac4537f9c4bf017"}},{"code_sha256_prefix":"456d16436e8dcb49","entry":"shell_command","repo":"mangopy/direct-rag-learning","repo_kind":"listed","path":"src/evaluate.py","file_url":"https://github.com/mangopy/direct-rag-learning/blob/HEAD/src/evaluate.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"456d16436e8dcb49"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}