{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:4ZEWPOPWCXBJH4AL65ZB5SUXIM","short_pith_number":"pith:4ZEWPOPW","schema_version":"1.0","canonical_sha256":"e64967b9f615c293f00bf7721eca97433b06eeecf75180b9877e5947bb672afb","source":{"kind":"arxiv","id":"2204.05212","version":2},"attestation_state":"computed","paper":{"title":"Single-Turn Debate Does Not Help Humans Answer Hard Reading-Comprehension Questions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alicia Parrish, Angelica Chen, Ethan Perez, Harsh Trivedi, Jason Phang, Nikita Nangia, Samuel R. Bowman","submitted_at":"2022-04-11T15:56:34Z","abstract_excerpt":"Current QA systems can generate reasonable-sounding yet false answers without explanation or evidence for the generated answer, which is especially problematic when humans cannot readily check the model's answers. This presents a challenge for building trust in machine learning systems. We take inspiration from real-world situations where difficult questions are answered by considering opposing sides (see Irving et al., 2018). For multiple-choice QA examples, we build a dataset of single arguments for both a correct and incorrect answer option in a debate-style set-up as an initial step in tra"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2204.05212","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-04-11T15:56:34Z","cross_cats_sorted":[],"title_canon_sha256":"b1f49107196bc2b14eacc58bceee6db343a392446f0967cf6f1fa07ebefc438b","abstract_canon_sha256":"7b686622ef6c0700098aa32aafcc9140d985191a75d9484cff86b69a1c6aa5c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:14:28.680767Z","signature_b64":"0v26uGZ4jioDQYvJcEZOKj8b2qMehG4Xt0DaSJO6rhqkk5q01Ony/NHjjBuXzjIEGJRUIXL4sFYHlZ3iufD7DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e64967b9f615c293f00bf7721eca97433b06eeecf75180b9877e5947bb672afb","last_reissued_at":"2026-07-05T04:14:28.680367Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:14:28.680367Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Single-Turn Debate Does Not Help Humans Answer Hard Reading-Comprehension Questions","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alicia Parrish, Angelica Chen, Ethan Perez, Harsh Trivedi, Jason Phang, Nikita Nangia, Samuel R. Bowman","submitted_at":"2022-04-11T15:56:34Z","abstract_excerpt":"Current QA systems can generate reasonable-sounding yet false answers without explanation or evidence for the generated answer, which is especially problematic when humans cannot readily check the model's answers. This presents a challenge for building trust in machine learning systems. We take inspiration from real-world situations where difficult questions are answered by considering opposing sides (see Irving et al., 2018). For multiple-choice QA examples, we build a dataset of single arguments for both a correct and incorrect answer option in a debate-style set-up as an initial step in tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.05212","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2204.05212/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2204.05212","created_at":"2026-07-05T04:14:28.680422+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.05212v2","created_at":"2026-07-05T04:14:28.680422+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.05212","created_at":"2026-07-05T04:14:28.680422+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ZEWPOPWCXBJ","created_at":"2026-07-05T04:14:28.680422+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ZEWPOPWCXBJH4AL","created_at":"2026-07-05T04:14:28.680422+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ZEWPOPW","created_at":"2026-07-05T04:14:28.680422+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.02866","citing_title":"When Helping Hurts and How to Fix It: Multi-Agent Debate for Data Cleaning","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2211.03540","citing_title":"Measuring Progress on Scalable Oversight for Large Language Models","ref_index":59,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM","json":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM.json","graph_json":"https://pith.science/api/pith-number/4ZEWPOPWCXBJH4AL65ZB5SUXIM/graph.json","events_json":"https://pith.science/api/pith-number/4ZEWPOPWCXBJH4AL65ZB5SUXIM/events.json","paper":"https://pith.science/paper/4ZEWPOPW"},"agent_actions":{"view_html":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM","download_json":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM.json","view_paper":"https://pith.science/paper/4ZEWPOPW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.05212&json=true","fetch_graph":"https://pith.science/api/pith-number/4ZEWPOPWCXBJH4AL65ZB5SUXIM/graph.json","fetch_events":"https://pith.science/api/pith-number/4ZEWPOPWCXBJH4AL65ZB5SUXIM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM/action/storage_attestation","attest_author":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM/action/author_attestation","sign_citation":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM/action/citation_signature","submit_replication":"https://pith.science/pith/4ZEWPOPWCXBJH4AL65ZB5SUXIM/action/replication_record"}},"created_at":"2026-07-05T04:14:28.680422+00:00","updated_at":"2026-07-05T04:14:28.680422+00:00"}