{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:EZSDCO2L5DLEHQODFUJBHMK4A3","short_pith_number":"pith:EZSDCO2L","canonical_record":{"source":{"id":"2112.08812","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T11:57:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e7462887c19783744eb6ba14ddb3ab02a084b6aba86f5d0fbc569f48150a6c3c","abstract_canon_sha256":"a8db266381cd597c879997c34171c5aafb559b7f15b451be2fd2bd70bc8d9d3b"},"schema_version":"1.0"},"canonical_sha256":"2664313b4be8d643c1c32d1213b15c06dfd77dab4b6d737c2a85d19d61b31950","source":{"kind":"arxiv","id":"2112.08812","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.08812","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"arxiv_version","alias_value":"2112.08812v2","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08812","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_12","alias_value":"EZSDCO2L5DLE","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_16","alias_value":"EZSDCO2L5DLEHQOD","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_8","alias_value":"EZSDCO2L","created_at":"2026-07-05T04:07:07Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:EZSDCO2L5DLEHQODFUJBHMK4A3","target":"record","payload":{"canonical_record":{"source":{"id":"2112.08812","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T11:57:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"e7462887c19783744eb6ba14ddb3ab02a084b6aba86f5d0fbc569f48150a6c3c","abstract_canon_sha256":"a8db266381cd597c879997c34171c5aafb559b7f15b451be2fd2bd70bc8d9d3b"},"schema_version":"1.0"},"canonical_sha256":"2664313b4be8d643c1c32d1213b15c06dfd77dab4b6d737c2a85d19d61b31950","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:07:07.022724Z","signature_b64":"NCB45xvca6bSR+XxTsBEkGQgTJ84aI5w9v06+ez4vEau8AMPlgjGKsOvfIxJrP79JFhJpbK0uYRz3YJS/zL3Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2664313b4be8d643c1c32d1213b15c06dfd77dab4b6d737c2a85d19d61b31950","last_reissued_at":"2026-07-05T04:07:07.022276Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:07:07.022276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.08812","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:07:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SPZBGuUDcrJg5T04U+YLTjlaOHz7+7Zvt5BbYACCgEAEAYZnYDYS6pr1yjN6VxxzjvaIXg4cHQQ9c90mbuDECA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:50:50.850656Z"},"content_sha256":"0dc51a238b8507d3e7bf7ce5445bca67d77f286e5888840a6b5b8e4073798bdf","schema_version":"1.0","event_id":"sha256:0dc51a238b8507d3e7bf7ce5445bca67d77f286e5888840a6b5b8e4073798bdf"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:EZSDCO2L5DLEHQODFUJBHMK4A3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Ditch the Gold Standard: Re-evaluating Conversational Question Answering","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Danqi Chen, Huihan Li, Manan Goenka, Tianyu Gao","submitted_at":"2021-12-16T11:57:56Z","abstract_excerpt":"Conversational question answering aims to provide natural-language answers to users in information-seeking conversations. Existing conversational QA benchmarks compare models with pre-collected human-human conversations, using ground-truth answers provided in conversational history. It remains unclear whether we can rely on this static evaluation for model development and whether current systems can well generalize to real-world human-machine conversations. In this work, we conduct the first large-scale human evaluation of state-of-the-art conversational QA systems, where human evaluators conv"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08812","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/2112.08812/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":"2026-07-05T04:07:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4++3HFLwQi/Ti7CX2978yQcDwA7n+0uywW0J//OMXolvS7vcyMc3l9iE38n0PyTfukkeSS2ER/XV6X9+cINMCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T05:50:50.851817Z"},"content_sha256":"ab1765f992802e33244eb4498e687910babab8e6af66792a0b0f80874fb5eac2","schema_version":"1.0","event_id":"sha256:ab1765f992802e33244eb4498e687910babab8e6af66792a0b0f80874fb5eac2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/bundle.json","state_url":"https://pith.science/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/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":"2026-08-09T05:50:50Z","links":{"resolver":"https://pith.science/pith/EZSDCO2L5DLEHQODFUJBHMK4A3","bundle":"https://pith.science/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/bundle.json","state":"https://pith.science/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EZSDCO2L5DLEHQODFUJBHMK4A3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:EZSDCO2L5DLEHQODFUJBHMK4A3","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":"a8db266381cd597c879997c34171c5aafb559b7f15b451be2fd2bd70bc8d9d3b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T11:57:56Z","title_canon_sha256":"e7462887c19783744eb6ba14ddb3ab02a084b6aba86f5d0fbc569f48150a6c3c"},"schema_version":"1.0","source":{"id":"2112.08812","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.08812","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"arxiv_version","alias_value":"2112.08812v2","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.08812","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_12","alias_value":"EZSDCO2L5DLE","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_16","alias_value":"EZSDCO2L5DLEHQOD","created_at":"2026-07-05T04:07:07Z"},{"alias_kind":"pith_short_8","alias_value":"EZSDCO2L","created_at":"2026-07-05T04:07:07Z"}],"graph_snapshots":[{"event_id":"sha256:ab1765f992802e33244eb4498e687910babab8e6af66792a0b0f80874fb5eac2","target":"graph","created_at":"2026-07-05T04:07:07Z","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/2112.08812/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Conversational question answering aims to provide natural-language answers to users in information-seeking conversations. Existing conversational QA benchmarks compare models with pre-collected human-human conversations, using ground-truth answers provided in conversational history. It remains unclear whether we can rely on this static evaluation for model development and whether current systems can well generalize to real-world human-machine conversations. In this work, we conduct the first large-scale human evaluation of state-of-the-art conversational QA systems, where human evaluators conv","authors_text":"Danqi Chen, Huihan Li, Manan Goenka, Tianyu Gao","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T11:57:56Z","title":"Ditch the Gold Standard: Re-evaluating Conversational Question Answering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.08812","kind":"arxiv","version":2},"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":"sha256:0dc51a238b8507d3e7bf7ce5445bca67d77f286e5888840a6b5b8e4073798bdf","target":"record","created_at":"2026-07-05T04:07:07Z","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":"a8db266381cd597c879997c34171c5aafb559b7f15b451be2fd2bd70bc8d9d3b","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-12-16T11:57:56Z","title_canon_sha256":"e7462887c19783744eb6ba14ddb3ab02a084b6aba86f5d0fbc569f48150a6c3c"},"schema_version":"1.0","source":{"id":"2112.08812","kind":"arxiv","version":2}},"canonical_sha256":"2664313b4be8d643c1c32d1213b15c06dfd77dab4b6d737c2a85d19d61b31950","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2664313b4be8d643c1c32d1213b15c06dfd77dab4b6d737c2a85d19d61b31950","first_computed_at":"2026-07-05T04:07:07.022276Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:07:07.022276Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NCB45xvca6bSR+XxTsBEkGQgTJ84aI5w9v06+ez4vEau8AMPlgjGKsOvfIxJrP79JFhJpbK0uYRz3YJS/zL3Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:07:07.022724Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.08812","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0dc51a238b8507d3e7bf7ce5445bca67d77f286e5888840a6b5b8e4073798bdf","sha256:ab1765f992802e33244eb4498e687910babab8e6af66792a0b0f80874fb5eac2"],"state_sha256":"c92bf37bf35112413522ac8f3d58964285dabc38cd372583db0fc304133ddf4f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vkdQz2r9m9MmFCO1pfNpgHojx7IZVBFXS43kfit51ISvGztcWeh70v9t7I7aExVr8YF6xr/8ab8meeeJfJOdAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T05:50:50.857530Z","bundle_sha256":"747872c4ad63adbe42dc11fdeadea054d7ac31a87787c8803f84a094d70caba7"}}