{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HT6QQ4TB7OGKVKI3UABOZZ4U73","short_pith_number":"pith:HT6QQ4TB","schema_version":"1.0","canonical_sha256":"3cfd087261fb8caaa91ba002ece794fec1c1fd4490c05decc7f17e6b34f67585","source":{"kind":"arxiv","id":"2403.07008","version":3},"attestation_state":"computed","paper":{"title":"AutoEval Done Right: Using Synthetic Data for Model Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ME"],"primary_cat":"cs.LG","authors_text":"Anastasios N. Angelopoulos, Jitendra Malik, Michael I. Jordan, Nir Yosef, Pierre Boyeau","submitted_at":"2024-03-09T02:47:11Z","abstract_excerpt":"The evaluation of machine learning models using human-labeled validation data can be expensive and time-consuming. AI-labeled synthetic data can be used to decrease the number of human annotations required for this purpose in a process called autoevaluation. We suggest efficient and statistically principled algorithms for this purpose that improve sample efficiency while remaining unbiased. These algorithms increase the effective human-labeled sample size by up to 50% on experiments with GPT-4."},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2403.07008","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-03-09T02:47:11Z","cross_cats_sorted":["cs.AI","cs.CL","stat.ME"],"title_canon_sha256":"4825a5bfdbb40bf4dd097eeeca06730d89bbabdeed7b64dcc922037f5fc7c59f","abstract_canon_sha256":"cfb29c4c262a64da1fc4447bc450c0e8515de1da77deb8084ab5abebae19d345"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:03.965864Z","signature_b64":"xyEHSvVB2uYZqTFA3dBpIPs53pyVMQk9/uUDCyXYWhcePMU2GM1dxe5GGwQEGOSDZyYd75yAuJRGrHF5BznjCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3cfd087261fb8caaa91ba002ece794fec1c1fd4490c05decc7f17e6b34f67585","last_reissued_at":"2026-06-02T02:04:03.965445Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:03.965445Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AutoEval Done Right: Using Synthetic Data for Model Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","stat.ME"],"primary_cat":"cs.LG","authors_text":"Anastasios N. Angelopoulos, Jitendra Malik, Michael I. Jordan, Nir Yosef, Pierre Boyeau","submitted_at":"2024-03-09T02:47:11Z","abstract_excerpt":"The evaluation of machine learning models using human-labeled validation data can be expensive and time-consuming. AI-labeled synthetic data can be used to decrease the number of human annotations required for this purpose in a process called autoevaluation. We suggest efficient and statistically principled algorithms for this purpose that improve sample efficiency while remaining unbiased. These algorithms increase the effective human-labeled sample size by up to 50% on experiments with GPT-4."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.07008","kind":"arxiv","version":3},"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/2403.07008/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2403.07008","created_at":"2026-06-02T02:04:03.965512+00:00"},{"alias_kind":"arxiv_version","alias_value":"2403.07008v3","created_at":"2026-06-02T02:04:03.965512+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.07008","created_at":"2026-06-02T02:04:03.965512+00:00"},{"alias_kind":"pith_short_12","alias_value":"HT6QQ4TB7OGK","created_at":"2026-06-02T02:04:03.965512+00:00"},{"alias_kind":"pith_short_16","alias_value":"HT6QQ4TB7OGKVKI3","created_at":"2026-06-02T02:04:03.965512+00:00"},{"alias_kind":"pith_short_8","alias_value":"HT6QQ4TB","created_at":"2026-06-02T02:04:03.965512+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73","json":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73.json","graph_json":"https://pith.science/api/pith-number/HT6QQ4TB7OGKVKI3UABOZZ4U73/graph.json","events_json":"https://pith.science/api/pith-number/HT6QQ4TB7OGKVKI3UABOZZ4U73/events.json","paper":"https://pith.science/paper/HT6QQ4TB"},"agent_actions":{"view_html":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73","download_json":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73.json","view_paper":"https://pith.science/paper/HT6QQ4TB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2403.07008&json=true","fetch_graph":"https://pith.science/api/pith-number/HT6QQ4TB7OGKVKI3UABOZZ4U73/graph.json","fetch_events":"https://pith.science/api/pith-number/HT6QQ4TB7OGKVKI3UABOZZ4U73/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73/action/storage_attestation","attest_author":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73/action/author_attestation","sign_citation":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73/action/citation_signature","submit_replication":"https://pith.science/pith/HT6QQ4TB7OGKVKI3UABOZZ4U73/action/replication_record"}},"created_at":"2026-06-02T02:04:03.965512+00:00","updated_at":"2026-06-02T02:04:03.965512+00:00"}