{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:VNS4EQHDGFKEJ3UWRELHYYDEWE","short_pith_number":"pith:VNS4EQHD","canonical_record":{"source":{"id":"2311.08702","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-15T05:05:40Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"3ded429e4b98ca792a6a7bb11fd1a0d1d5416ae3cdc2ed0366cdbf7938bc4358","abstract_canon_sha256":"a0e09171aafc16f3953e8b9fba1b0ea584fec727f2479df1208875a9c06426ba"},"schema_version":"1.0"},"canonical_sha256":"ab65c240e3315444ee9689167c6064b122a5ae9a39f5e62eda952cb872dd4a8c","source":{"kind":"arxiv","id":"2311.08702","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08702","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08702v1","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08702","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_12","alias_value":"VNS4EQHDGFKE","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_16","alias_value":"VNS4EQHDGFKEJ3UW","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_8","alias_value":"VNS4EQHD","created_at":"2026-07-05T07:13:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:VNS4EQHDGFKEJ3UWRELHYYDEWE","target":"record","payload":{"canonical_record":{"source":{"id":"2311.08702","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-15T05:05:40Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"3ded429e4b98ca792a6a7bb11fd1a0d1d5416ae3cdc2ed0366cdbf7938bc4358","abstract_canon_sha256":"a0e09171aafc16f3953e8b9fba1b0ea584fec727f2479df1208875a9c06426ba"},"schema_version":"1.0"},"canonical_sha256":"ab65c240e3315444ee9689167c6064b122a5ae9a39f5e62eda952cb872dd4a8c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:13:02.027832Z","signature_b64":"/BQJjjPhsVgh77fOFA4zb/PSEl+oqr/NkluYLQs6Ca74So+cAPV9elKb9iDmjoDQIKdMkHK5N8qWDW2ck263DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ab65c240e3315444ee9689167c6064b122a5ae9a39f5e62eda952cb872dd4a8c","last_reissued_at":"2026-07-05T07:13:02.027348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:13:02.027348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2311.08702","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:13:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"khrjqAmHogyEYFX9ZiayWq6OHdw1UOGT9a7/RgOlX+6Uf6C9K0lJyb75BVa0lcc2XEX36aT7LBCwDjQfdHBPAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:01:08.649847Z"},"content_sha256":"05b8f9ade58d287888d141add2414e4789eb09036c8f331debc5785cce591b77","schema_version":"1.0","event_id":"sha256:05b8f9ade58d287888d141add2414e4789eb09036c8f331debc5785cce591b77"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:VNS4EQHDGFKEJ3UWRELHYYDEWE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Debate Helps Supervise Unreliable Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"David Rein, Jackson Petty, Julian Michael, Julien Dirani, Salsabila Mahdi, Samuel R. Bowman, Vishakh Padmakumar","submitted_at":"2023-11-15T05:05:40Z","abstract_excerpt":"As AI systems are used to answer more difficult questions and potentially help create new knowledge, judging the truthfulness of their outputs becomes more difficult and more important. How can we supervise unreliable experts, which have access to the truth but may not accurately report it, to give answers that are systematically true and don't just superficially seem true, when the supervisor can't tell the difference between the two on their own? In this work, we show that debate between two unreliable experts can help a non-expert judge more reliably identify the truth. We collect a dataset"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08702","kind":"arxiv","version":1},"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/2311.08702/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:13:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AlChv91yyiofPOlNiDFIB5h/nhW9ZWPxIkNyLwXFIqNdRr4ljDyCm5RIdJ8V9Xkik8Ki8J4LamVCJakv6q4RDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T21:01:08.651769Z"},"content_sha256":"0812c1fbc31d4f46909b83b7a2156554f766e62f9822f333dbdef18c670c6486","schema_version":"1.0","event_id":"sha256:0812c1fbc31d4f46909b83b7a2156554f766e62f9822f333dbdef18c670c6486"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/bundle.json","state_url":"https://pith.science/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-08T21:01:08Z","links":{"resolver":"https://pith.science/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE","bundle":"https://pith.science/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/bundle.json","state":"https://pith.science/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VNS4EQHDGFKEJ3UWRELHYYDEWE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:VNS4EQHDGFKEJ3UWRELHYYDEWE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a0e09171aafc16f3953e8b9fba1b0ea584fec727f2479df1208875a9c06426ba","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-15T05:05:40Z","title_canon_sha256":"3ded429e4b98ca792a6a7bb11fd1a0d1d5416ae3cdc2ed0366cdbf7938bc4358"},"schema_version":"1.0","source":{"id":"2311.08702","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.08702","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"arxiv_version","alias_value":"2311.08702v1","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.08702","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_12","alias_value":"VNS4EQHDGFKE","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_16","alias_value":"VNS4EQHDGFKEJ3UW","created_at":"2026-07-05T07:13:02Z"},{"alias_kind":"pith_short_8","alias_value":"VNS4EQHD","created_at":"2026-07-05T07:13:02Z"}],"graph_snapshots":[{"event_id":"sha256:0812c1fbc31d4f46909b83b7a2156554f766e62f9822f333dbdef18c670c6486","target":"graph","created_at":"2026-07-05T07:13:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2311.08702/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As AI systems are used to answer more difficult questions and potentially help create new knowledge, judging the truthfulness of their outputs becomes more difficult and more important. How can we supervise unreliable experts, which have access to the truth but may not accurately report it, to give answers that are systematically true and don't just superficially seem true, when the supervisor can't tell the difference between the two on their own? In this work, we show that debate between two unreliable experts can help a non-expert judge more reliably identify the truth. We collect a dataset","authors_text":"David Rein, Jackson Petty, Julian Michael, Julien Dirani, Salsabila Mahdi, Samuel R. Bowman, Vishakh Padmakumar","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-15T05:05:40Z","title":"Debate Helps Supervise Unreliable Experts"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.08702","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:05b8f9ade58d287888d141add2414e4789eb09036c8f331debc5785cce591b77","target":"record","created_at":"2026-07-05T07:13:02Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a0e09171aafc16f3953e8b9fba1b0ea584fec727f2479df1208875a9c06426ba","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-11-15T05:05:40Z","title_canon_sha256":"3ded429e4b98ca792a6a7bb11fd1a0d1d5416ae3cdc2ed0366cdbf7938bc4358"},"schema_version":"1.0","source":{"id":"2311.08702","kind":"arxiv","version":1}},"canonical_sha256":"ab65c240e3315444ee9689167c6064b122a5ae9a39f5e62eda952cb872dd4a8c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ab65c240e3315444ee9689167c6064b122a5ae9a39f5e62eda952cb872dd4a8c","first_computed_at":"2026-07-05T07:13:02.027348Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:13:02.027348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/BQJjjPhsVgh77fOFA4zb/PSEl+oqr/NkluYLQs6Ca74So+cAPV9elKb9iDmjoDQIKdMkHK5N8qWDW2ck263DA==","signature_status":"signed_v1","signed_at":"2026-07-05T07:13:02.027832Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.08702","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05b8f9ade58d287888d141add2414e4789eb09036c8f331debc5785cce591b77","sha256:0812c1fbc31d4f46909b83b7a2156554f766e62f9822f333dbdef18c670c6486"],"state_sha256":"0f41484341a5bd0b405ba3f60574ce476dc5e533c53c1a15cb9095e9928a40d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"enAoW6biu32Rp4eDfIB0xoId/Yn0ObzEHTksretFFIMPKToNvEryj5VBhqVNaUTMn9Ml7kChVufGr7dsDesuDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T21:01:08.658614Z","bundle_sha256":"3d27ab38dd79682b80f1072588f08ef3f963f64e400c5895de251e504accdb77"}}