{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:EPLN2PRI6WQGNZ7SSCGHNTMGVD","short_pith_number":"pith:EPLN2PRI","schema_version":"1.0","canonical_sha256":"23d6dd3e28f5a066e7f2908c76cd86a8c93c538009a68f338b07daf00f6154d2","source":{"kind":"arxiv","id":"2405.20456","version":1},"attestation_state":"computed","paper":{"title":"Scaling Laws for the Value of Individual Data Points in Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ian Covert, James Zou, Tatsunori Hashimoto, Wenlong Ji","submitted_at":"2024-05-30T20:10:24Z","abstract_excerpt":"Recent works have shown that machine learning models improve at a predictable rate with the total amount of training data, leading to scaling laws that describe the relationship between error and dataset size. These scaling laws can help design a model's training dataset, but they typically take an aggregate view of the data by only considering the dataset's size. We introduce a new perspective by investigating scaling behavior for the value of individual data points: we find that a data point's contribution to model's performance shrinks predictably with the size of the dataset in a log-linea"},"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":"2405.20456","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-05-30T20:10:24Z","cross_cats_sorted":[],"title_canon_sha256":"53b1dbfcb2e483983713f6b02f4b841f2d9d0990226b126a593181fd863e28d0","abstract_canon_sha256":"fe6e05c93644edc31ddf6cdf3ef4d88f9470cb970330f4cd5f55572787467b64"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:25:37.779734Z","signature_b64":"3eZcKZc46jN9xCwxppG3RC+H8zUC6zWdSMHYjxqKMA4LSunB/aq3Rf6RF/7Xooex40F1ckm9yMFM7YyGlc9KBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"23d6dd3e28f5a066e7f2908c76cd86a8c93c538009a68f338b07daf00f6154d2","last_reissued_at":"2026-07-05T08:25:37.779260Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:25:37.779260Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Laws for the Value of Individual Data Points in Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ian Covert, James Zou, Tatsunori Hashimoto, Wenlong Ji","submitted_at":"2024-05-30T20:10:24Z","abstract_excerpt":"Recent works have shown that machine learning models improve at a predictable rate with the total amount of training data, leading to scaling laws that describe the relationship between error and dataset size. These scaling laws can help design a model's training dataset, but they typically take an aggregate view of the data by only considering the dataset's size. We introduce a new perspective by investigating scaling behavior for the value of individual data points: we find that a data point's contribution to model's performance shrinks predictably with the size of the dataset in a log-linea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20456","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/2405.20456/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":"2405.20456","created_at":"2026-07-05T08:25:37.779319+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.20456v1","created_at":"2026-07-05T08:25:37.779319+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20456","created_at":"2026-07-05T08:25:37.779319+00:00"},{"alias_kind":"pith_short_12","alias_value":"EPLN2PRI6WQG","created_at":"2026-07-05T08:25:37.779319+00:00"},{"alias_kind":"pith_short_16","alias_value":"EPLN2PRI6WQGNZ7S","created_at":"2026-07-05T08:25:37.779319+00:00"},{"alias_kind":"pith_short_8","alias_value":"EPLN2PRI","created_at":"2026-07-05T08:25:37.779319+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.15520","citing_title":"On the Fragility of Data Attribution When Learning Is Distributed","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD","json":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD.json","graph_json":"https://pith.science/api/pith-number/EPLN2PRI6WQGNZ7SSCGHNTMGVD/graph.json","events_json":"https://pith.science/api/pith-number/EPLN2PRI6WQGNZ7SSCGHNTMGVD/events.json","paper":"https://pith.science/paper/EPLN2PRI"},"agent_actions":{"view_html":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD","download_json":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD.json","view_paper":"https://pith.science/paper/EPLN2PRI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.20456&json=true","fetch_graph":"https://pith.science/api/pith-number/EPLN2PRI6WQGNZ7SSCGHNTMGVD/graph.json","fetch_events":"https://pith.science/api/pith-number/EPLN2PRI6WQGNZ7SSCGHNTMGVD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD/action/storage_attestation","attest_author":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD/action/author_attestation","sign_citation":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD/action/citation_signature","submit_replication":"https://pith.science/pith/EPLN2PRI6WQGNZ7SSCGHNTMGVD/action/replication_record"}},"created_at":"2026-07-05T08:25:37.779319+00:00","updated_at":"2026-07-05T08:25:37.779319+00:00"}