{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:5AUE5C3CZ3EBOR2MVGYVW57KOA","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":"303634eef574df5d52601015edcc42369b342419300cddb5c33321f37ba00006","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-30T19:00:25Z","title_canon_sha256":"54c254f22496843352f8e983bcfa3f43ee7a88e37122e15c48bb29d9f31edad9"},"schema_version":"1.0","source":{"id":"2107.00061","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.00061","created_at":"2026-07-05T02:55:56Z"},{"alias_kind":"arxiv_version","alias_value":"2107.00061v2","created_at":"2026-07-05T02:55:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.00061","created_at":"2026-07-05T02:55:56Z"},{"alias_kind":"pith_short_12","alias_value":"5AUE5C3CZ3EB","created_at":"2026-07-05T02:55:56Z"},{"alias_kind":"pith_short_16","alias_value":"5AUE5C3CZ3EBOR2M","created_at":"2026-07-05T02:55:56Z"},{"alias_kind":"pith_short_8","alias_value":"5AUE5C3C","created_at":"2026-07-05T02:55:56Z"}],"graph_snapshots":[{"event_id":"sha256:97abb5ad7a9ce79c61d39ce2358e652a842872fba0ee6c286df5e4cad3622e6c","target":"graph","created_at":"2026-07-05T02:55:56Z","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/2107.00061/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Human evaluations are typically considered the gold standard in natural language generation, but as models' fluency improves, how well can evaluators detect and judge machine-generated text? We run a study assessing non-experts' ability to distinguish between human- and machine-authored text (GPT2 and GPT3) in three domains (stories, news articles, and recipes). We find that, without training, evaluators distinguished between GPT3- and human-authored text at random chance level. We explore three approaches for quickly training evaluators to better identify GPT3-authored text (detailed instruct","authors_text":"Elizabeth Clark, Nikita Haduong, Noah A. Smith, Sofia Serrano, Suchin Gururangan, Tal August","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-30T19:00:25Z","title":"All That's 'Human' Is Not Gold: Evaluating Human Evaluation of Generated Text"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.00061","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":"sha256:83d8bcb0b0307ae413655467805f1108cf8888e009bd14850b2ca9807e56f7d3","target":"record","created_at":"2026-07-05T02:55:56Z","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":"303634eef574df5d52601015edcc42369b342419300cddb5c33321f37ba00006","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-30T19:00:25Z","title_canon_sha256":"54c254f22496843352f8e983bcfa3f43ee7a88e37122e15c48bb29d9f31edad9"},"schema_version":"1.0","source":{"id":"2107.00061","kind":"arxiv","version":2}},"canonical_sha256":"e8284e8b62cec817474ca9b15b77ea7009f22b650b1bbc7d666f4bf9ca0688b0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8284e8b62cec817474ca9b15b77ea7009f22b650b1bbc7d666f4bf9ca0688b0","first_computed_at":"2026-07-05T02:55:56.456148Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:55:56.456148Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ow6o3p1OPROI4QB2ulgiJO4fjsuDJEl1d2XyvtHe363W6jXaaBmoOxTUQPz0garHTHhIlXRJtJFVxIJGRkvtDg==","signature_status":"signed_v1","signed_at":"2026-07-05T02:55:56.456583Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.00061","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:83d8bcb0b0307ae413655467805f1108cf8888e009bd14850b2ca9807e56f7d3","sha256:97abb5ad7a9ce79c61d39ce2358e652a842872fba0ee6c286df5e4cad3622e6c"],"state_sha256":"5a401e1e8caf74401cbc509b6f696a82827cc5ede8da529754baf7e2a1b1f17a"}