{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:WAY6RWYNQPTBNHYBZL4WX26OUY","short_pith_number":"pith:WAY6RWYN","canonical_record":{"source":{"id":"2011.03992","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-08T14:49:18Z","cross_cats_sorted":[],"title_canon_sha256":"8d801e0d8a360acc00446d2d44980db5c999d4cad84bd39e96ec9cc9062b0fcb","abstract_canon_sha256":"3b322b77b7a7f8ff69525661238ef74eb457c64847bfc82669510ab539d9730a"},"schema_version":"1.0"},"canonical_sha256":"b031e8db0d83e6169f01caf96bebcea607a11a65766eb0189d2d747f322d049a","source":{"kind":"arxiv","id":"2011.03992","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2011.03992","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"arxiv_version","alias_value":"2011.03992v1","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.03992","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_12","alias_value":"WAY6RWYNQPTB","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_16","alias_value":"WAY6RWYNQPTBNHYB","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_8","alias_value":"WAY6RWYN","created_at":"2026-07-05T01:50:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:WAY6RWYNQPTBNHYBZL4WX26OUY","target":"record","payload":{"canonical_record":{"source":{"id":"2011.03992","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-08T14:49:18Z","cross_cats_sorted":[],"title_canon_sha256":"8d801e0d8a360acc00446d2d44980db5c999d4cad84bd39e96ec9cc9062b0fcb","abstract_canon_sha256":"3b322b77b7a7f8ff69525661238ef74eb457c64847bfc82669510ab539d9730a"},"schema_version":"1.0"},"canonical_sha256":"b031e8db0d83e6169f01caf96bebcea607a11a65766eb0189d2d747f322d049a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:50:01.425184Z","signature_b64":"0z/DxMT2XShQTRKDTHq0zI1dP5+b6dIqwYWrdqZ/mIh+ZEVo4U2GFhpJElKqlFCl1HprGusSVQ/2FuAqOZ1HAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b031e8db0d83e6169f01caf96bebcea607a11a65766eb0189d2d747f322d049a","last_reissued_at":"2026-07-05T01:50:01.424795Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:50:01.424795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2011.03992","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":"2026-07-05T01:50:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bbKMJKvtimSx2XZw5VJD3iBdGk0Ixgqqz8mIm+idGZQAt7ZwScuECDzHRiqs7FHWDREy2eyJc5GqyeXergQECw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:09:17.176995Z"},"content_sha256":"fb4eae44dff2cecdbdc6a488e69c06567432d154f5fea5825c926a0feea654c6","schema_version":"1.0","event_id":"sha256:fb4eae44dff2cecdbdc6a488e69c06567432d154f5fea5825c926a0feea654c6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:WAY6RWYNQPTBNHYBZL4WX26OUY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A Gold Standard Methodology for Evaluating Accuracy in Data-To-Text Systems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Craig Thomson, Ehud Reiter","submitted_at":"2020-11-08T14:49:18Z","abstract_excerpt":"Most Natural Language Generation systems need to produce accurate texts. We propose a methodology for high-quality human evaluation of the accuracy of generated texts, which is intended to serve as a gold-standard for accuracy evaluations of data-to-text systems. We use our methodology to evaluate the accuracy of computer generated basketball summaries. We then show how our gold standard evaluation can be used to validate automated metrics"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.03992","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/2011.03992/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":"2026-07-05T01:50:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hIWSdaJ0vhjk/QnzdINVCfC9bASSkspY9uzTPeBKBtXEMKn3JTkk/Dq1OSkYlT4Jfa+jr3KvBi9XnWQn9GjKAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T17:09:17.177510Z"},"content_sha256":"083015f629f7f5d30f67cef2854159e10b99cf6526689a94cad1fca2e8e50868","schema_version":"1.0","event_id":"sha256:083015f629f7f5d30f67cef2854159e10b99cf6526689a94cad1fca2e8e50868"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/bundle.json","state_url":"https://pith.science/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/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":"2026-08-03T17:09:17Z","links":{"resolver":"https://pith.science/pith/WAY6RWYNQPTBNHYBZL4WX26OUY","bundle":"https://pith.science/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/bundle.json","state":"https://pith.science/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WAY6RWYNQPTBNHYBZL4WX26OUY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:WAY6RWYNQPTBNHYBZL4WX26OUY","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":"3b322b77b7a7f8ff69525661238ef74eb457c64847bfc82669510ab539d9730a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-08T14:49:18Z","title_canon_sha256":"8d801e0d8a360acc00446d2d44980db5c999d4cad84bd39e96ec9cc9062b0fcb"},"schema_version":"1.0","source":{"id":"2011.03992","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2011.03992","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"arxiv_version","alias_value":"2011.03992v1","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2011.03992","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_12","alias_value":"WAY6RWYNQPTB","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_16","alias_value":"WAY6RWYNQPTBNHYB","created_at":"2026-07-05T01:50:01Z"},{"alias_kind":"pith_short_8","alias_value":"WAY6RWYN","created_at":"2026-07-05T01:50:01Z"}],"graph_snapshots":[{"event_id":"sha256:083015f629f7f5d30f67cef2854159e10b99cf6526689a94cad1fca2e8e50868","target":"graph","created_at":"2026-07-05T01:50:01Z","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/2011.03992/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Most Natural Language Generation systems need to produce accurate texts. We propose a methodology for high-quality human evaluation of the accuracy of generated texts, which is intended to serve as a gold-standard for accuracy evaluations of data-to-text systems. We use our methodology to evaluate the accuracy of computer generated basketball summaries. We then show how our gold standard evaluation can be used to validate automated metrics","authors_text":"Craig Thomson, Ehud Reiter","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-08T14:49:18Z","title":"A Gold Standard Methodology for Evaluating Accuracy in Data-To-Text Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2011.03992","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:fb4eae44dff2cecdbdc6a488e69c06567432d154f5fea5825c926a0feea654c6","target":"record","created_at":"2026-07-05T01:50:01Z","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":"3b322b77b7a7f8ff69525661238ef74eb457c64847bfc82669510ab539d9730a","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2020-11-08T14:49:18Z","title_canon_sha256":"8d801e0d8a360acc00446d2d44980db5c999d4cad84bd39e96ec9cc9062b0fcb"},"schema_version":"1.0","source":{"id":"2011.03992","kind":"arxiv","version":1}},"canonical_sha256":"b031e8db0d83e6169f01caf96bebcea607a11a65766eb0189d2d747f322d049a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b031e8db0d83e6169f01caf96bebcea607a11a65766eb0189d2d747f322d049a","first_computed_at":"2026-07-05T01:50:01.424795Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:50:01.424795Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0z/DxMT2XShQTRKDTHq0zI1dP5+b6dIqwYWrdqZ/mIh+ZEVo4U2GFhpJElKqlFCl1HprGusSVQ/2FuAqOZ1HAA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:50:01.425184Z","signed_message":"canonical_sha256_bytes"},"source_id":"2011.03992","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fb4eae44dff2cecdbdc6a488e69c06567432d154f5fea5825c926a0feea654c6","sha256:083015f629f7f5d30f67cef2854159e10b99cf6526689a94cad1fca2e8e50868"],"state_sha256":"2a1334029e5e01ab4a4c6b725ead658a93011ce8e2cc820b8a040a0693d29b39"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lXbzbqHaGFpOKsCLqOaPOBdoeOmbhvcQuTMVVd2xrN7swOMBk8JeOiOZ2SnsqNMGa0uceaUXsw+LAWwYaiipCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T17:09:17.181927Z","bundle_sha256":"b37fd834e1d8132cb5659c42d923531894f58ffb9a01aebbf81258646b53fd5b"}}