{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:XWP5HQZ2JGBUWMZV7UINV6QLY6","short_pith_number":"pith:XWP5HQZ2","canonical_record":{"source":{"id":"2106.01494","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T22:22:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3af89320f854be7041325824e0676561a93fed1658543b9bb632c08b7dae8414","abstract_canon_sha256":"96a1dc3ffb1837f000bbe40543bb627e8c6f42c5a02c2898014c38dd53244829"},"schema_version":"1.0"},"canonical_sha256":"bd9fd3c33a49834b3335fd10dafa0bc7abbfa95f50c47d9256852067e64aea90","source":{"kind":"arxiv","id":"2106.01494","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.01494","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"arxiv_version","alias_value":"2106.01494v1","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01494","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_12","alias_value":"XWP5HQZ2JGBU","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_16","alias_value":"XWP5HQZ2JGBUWMZV","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_8","alias_value":"XWP5HQZ2","created_at":"2026-07-05T02:45:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:XWP5HQZ2JGBUWMZV7UINV6QLY6","target":"record","payload":{"canonical_record":{"source":{"id":"2106.01494","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T22:22:52Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3af89320f854be7041325824e0676561a93fed1658543b9bb632c08b7dae8414","abstract_canon_sha256":"96a1dc3ffb1837f000bbe40543bb627e8c6f42c5a02c2898014c38dd53244829"},"schema_version":"1.0"},"canonical_sha256":"bd9fd3c33a49834b3335fd10dafa0bc7abbfa95f50c47d9256852067e64aea90","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:45:52.129815Z","signature_b64":"bfUhp6xZcheZHaMb40s7BviPrIqMnue6KpFQt4JPlbHiasi4Kgy9d8zsqgdNoCqxaaQqwMi1RAKHePr6qgB9DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bd9fd3c33a49834b3335fd10dafa0bc7abbfa95f50c47d9256852067e64aea90","last_reissued_at":"2026-07-05T02:45:52.129451Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:45:52.129451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2106.01494","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":"2026-07-05T02:45:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JKJAHR3C2A8iU2UeMt+hpoSaM08A/B5URCR2BXBW7VWNESd2fmUePJHwdizwleHqp6sI8mE72QUF1FSrn/NfDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T13:00:15.008932Z"},"content_sha256":"d5b79c64bc2b1c1dea4de5cd8495157e891926d29208060d874b53ee4717c6f3","schema_version":"1.0","event_id":"sha256:d5b79c64bc2b1c1dea4de5cd8495157e891926d29208060d874b53ee4717c6f3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:XWP5HQZ2JGBUWMZV7UINV6QLY6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Knowing More About Questions Can Help: Improving Calibration in Question Answering","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chengyue Gong, Eunsol Choi, Shujian Zhang","submitted_at":"2021-06-02T22:22:52Z","abstract_excerpt":"We study calibration in question answering, estimating whether model correctly predicts answer for each question. Unlike prior work which mainly rely on the model's confidence score, our calibrator incorporates information about the input example (e.g., question and the evidence context). Together with data augmentation via back translation, our simple approach achieves 5-10% gains in calibration accuracy on reading comprehension benchmarks. Furthermore, we present the first calibration study in the open retrieval setting, comparing the calibration accuracy of retrieval-based span prediction m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01494","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/2106.01494/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":"2026-07-05T02:45:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bF/79aHFQlH9FxGllKknCDlB1248atY4kNcos8/ah9U0f6PV0KYFQQMhaZXjFI0DDh2zanY4xt+/9pSbCmgEAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T13:00:15.009818Z"},"content_sha256":"f4753982c3895785f2f33227daa5406be4926646df0c6d0afe33e330699e4db9","schema_version":"1.0","event_id":"sha256:f4753982c3895785f2f33227daa5406be4926646df0c6d0afe33e330699e4db9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/bundle.json","state_url":"https://pith.science/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/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":"2026-08-14T13:00:15Z","links":{"resolver":"https://pith.science/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6","bundle":"https://pith.science/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/bundle.json","state":"https://pith.science/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XWP5HQZ2JGBUWMZV7UINV6QLY6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:XWP5HQZ2JGBUWMZV7UINV6QLY6","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":"96a1dc3ffb1837f000bbe40543bb627e8c6f42c5a02c2898014c38dd53244829","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T22:22:52Z","title_canon_sha256":"3af89320f854be7041325824e0676561a93fed1658543b9bb632c08b7dae8414"},"schema_version":"1.0","source":{"id":"2106.01494","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.01494","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"arxiv_version","alias_value":"2106.01494v1","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.01494","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_12","alias_value":"XWP5HQZ2JGBU","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_16","alias_value":"XWP5HQZ2JGBUWMZV","created_at":"2026-07-05T02:45:52Z"},{"alias_kind":"pith_short_8","alias_value":"XWP5HQZ2","created_at":"2026-07-05T02:45:52Z"}],"graph_snapshots":[{"event_id":"sha256:f4753982c3895785f2f33227daa5406be4926646df0c6d0afe33e330699e4db9","target":"graph","created_at":"2026-07-05T02:45:52Z","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/2106.01494/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study calibration in question answering, estimating whether model correctly predicts answer for each question. Unlike prior work which mainly rely on the model's confidence score, our calibrator incorporates information about the input example (e.g., question and the evidence context). Together with data augmentation via back translation, our simple approach achieves 5-10% gains in calibration accuracy on reading comprehension benchmarks. Furthermore, we present the first calibration study in the open retrieval setting, comparing the calibration accuracy of retrieval-based span prediction m","authors_text":"Chengyue Gong, Eunsol Choi, Shujian Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T22:22:52Z","title":"Knowing More About Questions Can Help: Improving Calibration in Question Answering"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.01494","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:d5b79c64bc2b1c1dea4de5cd8495157e891926d29208060d874b53ee4717c6f3","target":"record","created_at":"2026-07-05T02:45:52Z","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":"96a1dc3ffb1837f000bbe40543bb627e8c6f42c5a02c2898014c38dd53244829","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-02T22:22:52Z","title_canon_sha256":"3af89320f854be7041325824e0676561a93fed1658543b9bb632c08b7dae8414"},"schema_version":"1.0","source":{"id":"2106.01494","kind":"arxiv","version":1}},"canonical_sha256":"bd9fd3c33a49834b3335fd10dafa0bc7abbfa95f50c47d9256852067e64aea90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bd9fd3c33a49834b3335fd10dafa0bc7abbfa95f50c47d9256852067e64aea90","first_computed_at":"2026-07-05T02:45:52.129451Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:45:52.129451Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bfUhp6xZcheZHaMb40s7BviPrIqMnue6KpFQt4JPlbHiasi4Kgy9d8zsqgdNoCqxaaQqwMi1RAKHePr6qgB9DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:45:52.129815Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.01494","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d5b79c64bc2b1c1dea4de5cd8495157e891926d29208060d874b53ee4717c6f3","sha256:f4753982c3895785f2f33227daa5406be4926646df0c6d0afe33e330699e4db9"],"state_sha256":"1ab2d505b04dba198a4e3b914ccc30fc5292cdcb718fd9f1ae7d1ff1e3a91f80"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SX2VTYgFOeWAbjzNxd/QSivFL2GkePO7ABeDhk6fcveAD9K6s1YY2nZINis1ozP7xP9OqOPhU8c3SL1k/Ei3AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T13:00:15.017202Z","bundle_sha256":"3d8e36abf8210ac2be830c78c4ba31e32d75b66fec343f20774310833bfaeef3"}}