{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GACU7GHBFS4R3QDSSXWKS7AKM7","short_pith_number":"pith:GACU7GHB","schema_version":"1.0","canonical_sha256":"30054f98e12cb91dc07295eca97c0a67fcaa6727f7b57d89efd552a2e7dc62a9","source":{"kind":"arxiv","id":"2601.20819","version":2},"attestation_state":"computed","paper":{"title":"Demystifying Prediction Powered Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Dan M. Kluger, Harsh Parikh, Tian Gu, Yilin Song","submitted_at":"2026-01-28T18:16:02Z","abstract_excerpt":"Machine learning predictions are increasingly used to supplement incomplete or costly-to-measure outcomes in fields such as biomedical research, environmental science, and social science. However, treating predictions as ground truth introduces bias while ignoring them wastes valuable information. Prediction-Powered Inference (PPI) offers a principled framework that leverages predictions from large unlabeled datasets to improve statistical efficiency while maintaining valid inference through explicit bias correction using a smaller labeled subset. Despite its potential, the growing PPI variant"},"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":"2601.20819","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-01-28T18:16:02Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2a41fa5f40536cc4fca12cb99a99c3c6a2189d83e0236cef190a4ae87f09bfbf","abstract_canon_sha256":"b447a52b7d15b65d3684bdf9a1ee6d2d59748ee58c06e8c5e47c2d322c5acdd7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-11T01:20:25.227590Z","signature_b64":"j5BurRhV/BRd4v/4mu3Zccg4911YFjOvbK7SpIBrUI7PlatN2PXb6Y/uklmmxwQQ2+sjyHX/TK8RGz3pSi4ABA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30054f98e12cb91dc07295eca97c0a67fcaa6727f7b57d89efd552a2e7dc62a9","last_reissued_at":"2026-08-11T01:20:25.224484Z","signature_status":"signed_v1","first_computed_at":"2026-08-11T01:20:25.224484Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Demystifying Prediction Powered Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Dan M. Kluger, Harsh Parikh, Tian Gu, Yilin Song","submitted_at":"2026-01-28T18:16:02Z","abstract_excerpt":"Machine learning predictions are increasingly used to supplement incomplete or costly-to-measure outcomes in fields such as biomedical research, environmental science, and social science. However, treating predictions as ground truth introduces bias while ignoring them wastes valuable information. Prediction-Powered Inference (PPI) offers a principled framework that leverages predictions from large unlabeled datasets to improve statistical efficiency while maintaining valid inference through explicit bias correction using a smaller labeled subset. Despite its potential, the growing PPI variant"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.20819","kind":"arxiv","version":2},"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/2601.20819/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":"2601.20819","created_at":"2026-08-11T01:20:25.225507+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.20819v2","created_at":"2026-08-11T01:20:25.225507+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.20819","created_at":"2026-08-11T01:20:25.225507+00:00"},{"alias_kind":"pith_short_12","alias_value":"GACU7GHBFS4R","created_at":"2026-08-11T01:20:25.225507+00:00"},{"alias_kind":"pith_short_16","alias_value":"GACU7GHBFS4R3QDS","created_at":"2026-08-11T01:20:25.225507+00:00"},{"alias_kind":"pith_short_8","alias_value":"GACU7GHB","created_at":"2026-08-11T01:20:25.225507+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":7,"sample":[{"citing_arxiv_id":"2606.25601","citing_title":"Statistically Valid Hyperparameter Selection: From Tuning to Guarantees","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2606.13629","citing_title":"Valid Inference with Synthetic Data via Task Exchangeability","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2606.08853","citing_title":"AI-Assisted Variance Reduction in Randomized Experiments","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2605.29249","citing_title":"Prediction-Powered Inference Across Many Tasks for AI Evaluation & Social Science Research","ref_index":9,"is_internal_anchor":true},{"citing_arxiv_id":"2605.31278","citing_title":"Industrializing Prediction-Powered Inference: The GLIDE Library for Reliable GenAI and Agentic Systems Evaluation","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09351","citing_title":"In-Context Learning for the Imputation of Public Opinion Data with Large Language Models","ref_index":38,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21260","citing_title":"Calibeating Prediction-Powered Inference","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7","json":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7.json","graph_json":"https://pith.science/api/pith-number/GACU7GHBFS4R3QDSSXWKS7AKM7/graph.json","events_json":"https://pith.science/api/pith-number/GACU7GHBFS4R3QDSSXWKS7AKM7/events.json","paper":"https://pith.science/paper/GACU7GHB"},"agent_actions":{"view_html":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7","download_json":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7.json","view_paper":"https://pith.science/paper/GACU7GHB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.20819&json=true","fetch_graph":"https://pith.science/api/pith-number/GACU7GHBFS4R3QDSSXWKS7AKM7/graph.json","fetch_events":"https://pith.science/api/pith-number/GACU7GHBFS4R3QDSSXWKS7AKM7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7/action/storage_attestation","attest_author":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7/action/author_attestation","sign_citation":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7/action/citation_signature","submit_replication":"https://pith.science/pith/GACU7GHBFS4R3QDSSXWKS7AKM7/action/replication_record"}},"created_at":"2026-08-11T01:20:25.225507+00:00","updated_at":"2026-08-11T01:20:25.225507+00:00"}