{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:FSFSXM5FYWTQ6NAVWEAB4OU6FD","short_pith_number":"pith:FSFSXM5F","canonical_record":{"source":{"id":"1904.02792","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-04T21:03:34Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"6b3bf31fb490162a5b85bc68bdfedbeacb40dc415ce98e3c8ca7679cdc7d9018","abstract_canon_sha256":"c773e9c23a5346c8823b69a905b3f5891b1dd7cbfe80ddcb92e8d124c41e8abc"},"schema_version":"1.0"},"canonical_sha256":"2c8b2bb3a5c5a70f3415b1001e3a9e28d8bd989219ca96736bca4306351f003a","source":{"kind":"arxiv","id":"1904.02792","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.02792","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"arxiv_version","alias_value":"1904.02792v1","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.02792","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"pith_short_12","alias_value":"FSFSXM5FYWTQ","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"FSFSXM5FYWTQ6NAV","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"FSFSXM5F","created_at":"2026-05-18T12:33:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:FSFSXM5FYWTQ6NAVWEAB4OU6FD","target":"record","payload":{"canonical_record":{"source":{"id":"1904.02792","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-04T21:03:34Z","cross_cats_sorted":["cs.AI","stat.ML"],"title_canon_sha256":"6b3bf31fb490162a5b85bc68bdfedbeacb40dc415ce98e3c8ca7679cdc7d9018","abstract_canon_sha256":"c773e9c23a5346c8823b69a905b3f5891b1dd7cbfe80ddcb92e8d124c41e8abc"},"schema_version":"1.0"},"canonical_sha256":"2c8b2bb3a5c5a70f3415b1001e3a9e28d8bd989219ca96736bca4306351f003a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:19.628508Z","signature_b64":"LyGb53uG7qmlpKrqzoRTTzobm4/9IcKnInyqcQoVeiPLXUr3MwrkGViXfX2HIlT1DTDPRgAkaptvzh9U7NwMAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c8b2bb3a5c5a70f3415b1001e3a9e28d8bd989219ca96736bca4306351f003a","last_reissued_at":"2026-05-17T23:49:19.628056Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:19.628056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1904.02792","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":"2026-05-17T23:49:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ykT2v6ez1ZAkGzRLPkP6MY67/cWzIKxwnGbGkWdH2c2J/EgzVpMO2lepnGXVNrid3vL8ie5GF8dY19qlr5A9AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T07:11:43.088976Z"},"content_sha256":"dc2091373c407660e3e67846793c110ba3fa41765586a9c746cb676ca230a470","schema_version":"1.0","event_id":"sha256:dc2091373c407660e3e67846793c110ba3fa41765586a9c746cb676ca230a470"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:FSFSXM5FYWTQ6NAVWEAB4OU6FD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Unifying Human and Statistical Evaluation for Natural Language Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","stat.ML"],"primary_cat":"cs.CL","authors_text":"Hugh Zhang, Percy Liang, Tatsunori B. Hashimoto","submitted_at":"2019-04-04T21:03:34Z","abstract_excerpt":"How can we measure whether a natural language generation system produces both high quality and diverse outputs? Human evaluation captures quality but not diversity, as it does not catch models that simply plagiarize from the training set. On the other hand, statistical evaluation (i.e., perplexity) captures diversity but not quality, as models that occasionally emit low quality samples would be insufficiently penalized. In this paper, we propose a unified framework which evaluates both diversity and quality, based on the optimal error rate of predicting whether a sentence is human- or machine-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.02792","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":""},"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":"2026-05-17T23:49:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"q7RUSTHVfefriIN4zYgLPQxWUjUP27nnQoOwXSBopyOL76Tq1uIacbEKSbS/gEfeiIWnU73aiqwY3E71bOjwCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-10T07:11:43.089764Z"},"content_sha256":"05932f5ea18c790d0765ef3529ab7a735c2df39f83b604ad44b3e7e08350741c","schema_version":"1.0","event_id":"sha256:05932f5ea18c790d0765ef3529ab7a735c2df39f83b604ad44b3e7e08350741c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/bundle.json","state_url":"https://pith.science/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/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":"2026-06-10T07:11:43Z","links":{"resolver":"https://pith.science/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD","bundle":"https://pith.science/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/bundle.json","state":"https://pith.science/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FSFSXM5FYWTQ6NAVWEAB4OU6FD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:FSFSXM5FYWTQ6NAVWEAB4OU6FD","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":"c773e9c23a5346c8823b69a905b3f5891b1dd7cbfe80ddcb92e8d124c41e8abc","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-04T21:03:34Z","title_canon_sha256":"6b3bf31fb490162a5b85bc68bdfedbeacb40dc415ce98e3c8ca7679cdc7d9018"},"schema_version":"1.0","source":{"id":"1904.02792","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1904.02792","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"arxiv_version","alias_value":"1904.02792v1","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1904.02792","created_at":"2026-05-17T23:49:19Z"},{"alias_kind":"pith_short_12","alias_value":"FSFSXM5FYWTQ","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_16","alias_value":"FSFSXM5FYWTQ6NAV","created_at":"2026-05-18T12:33:15Z"},{"alias_kind":"pith_short_8","alias_value":"FSFSXM5F","created_at":"2026-05-18T12:33:15Z"}],"graph_snapshots":[{"event_id":"sha256:05932f5ea18c790d0765ef3529ab7a735c2df39f83b604ad44b3e7e08350741c","target":"graph","created_at":"2026-05-17T23:49:19Z","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"},"paper":{"abstract_excerpt":"How can we measure whether a natural language generation system produces both high quality and diverse outputs? Human evaluation captures quality but not diversity, as it does not catch models that simply plagiarize from the training set. On the other hand, statistical evaluation (i.e., perplexity) captures diversity but not quality, as models that occasionally emit low quality samples would be insufficiently penalized. In this paper, we propose a unified framework which evaluates both diversity and quality, based on the optimal error rate of predicting whether a sentence is human- or machine-","authors_text":"Hugh Zhang, Percy Liang, Tatsunori B. Hashimoto","cross_cats":["cs.AI","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-04T21:03:34Z","title":"Unifying Human and Statistical Evaluation for Natural Language Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1904.02792","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:dc2091373c407660e3e67846793c110ba3fa41765586a9c746cb676ca230a470","target":"record","created_at":"2026-05-17T23:49:19Z","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":"c773e9c23a5346c8823b69a905b3f5891b1dd7cbfe80ddcb92e8d124c41e8abc","cross_cats_sorted":["cs.AI","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-04-04T21:03:34Z","title_canon_sha256":"6b3bf31fb490162a5b85bc68bdfedbeacb40dc415ce98e3c8ca7679cdc7d9018"},"schema_version":"1.0","source":{"id":"1904.02792","kind":"arxiv","version":1}},"canonical_sha256":"2c8b2bb3a5c5a70f3415b1001e3a9e28d8bd989219ca96736bca4306351f003a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2c8b2bb3a5c5a70f3415b1001e3a9e28d8bd989219ca96736bca4306351f003a","first_computed_at":"2026-05-17T23:49:19.628056Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:49:19.628056Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"LyGb53uG7qmlpKrqzoRTTzobm4/9IcKnInyqcQoVeiPLXUr3MwrkGViXfX2HIlT1DTDPRgAkaptvzh9U7NwMAQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:49:19.628508Z","signed_message":"canonical_sha256_bytes"},"source_id":"1904.02792","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dc2091373c407660e3e67846793c110ba3fa41765586a9c746cb676ca230a470","sha256:05932f5ea18c790d0765ef3529ab7a735c2df39f83b604ad44b3e7e08350741c"],"state_sha256":"38f1cfcff62e9eedbd6fe4c1b578dc6c900efd5959253e80bc515cba333a9833"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qTH1l2d1wv9cvTgbywW/LeAKR4a8EJcevfzEXoTGz/1bKFZaZ523lTFyOtDBsZg5t6DxNvEvkQ76iR1GNxUQDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-10T07:11:43.093742Z","bundle_sha256":"5131540ff42707925e53ed20e8e0ba1986e0b391e461e0c3d0bfeb6b6ccb28d0"}}