{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:QVCTTPUVZG5AE3KRA4FQWE6PHB","short_pith_number":"pith:QVCTTPUV","schema_version":"1.0","canonical_sha256":"854539be95c9ba026d51070b0b13cf3845db4b6d8936f37ca0003c21a4b3eb99","source":{"kind":"arxiv","id":"2607.22766","version":1},"attestation_state":"computed","paper":{"title":"Beyond Shapley: An Influence-Based Data Auditing Pipeline for LLM Alignment and Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jun Qin, Matthew Watson, Peter Grabowski, Yunting Song","submitted_at":"2026-07-24T03:31:44Z","abstract_excerpt":"The alignment of Large Language Models (LLMs) is increasingly bottlenecked by data quality. As datasets scale, massive preference and instruction-tuning corpora inevitably accumulate hidden structural contradictions, safety risks, and systemic human annotation errors. Standard dataset auditing methods, such as semantic deduplication or LLM-as-a-judge, struggle to capture the actual predictive impact of individual records and often miss deep functional rule clashes. To address this, we introduce a scalable, inference-only data valuation pipeline that approximates the Shapley value without itera"},"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":"2607.22766","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-24T03:31:44Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"17df459e1cd9de46eb36bc9b65462d2e48b9aba141297d90ec1b3fbc56f81da0","abstract_canon_sha256":"a6a1071f055cb542af4df283acd76f98da44c33a91612754e74b1ef23274453c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T00:21:53.346969Z","signature_b64":"KxL0p4/aYPwkUYhcyjC4wpvRqVy5J4ck+M+QcY+7HQqsqdGXwJpoY+6khYbQTmC2In/FDi0KSD9oI0XFMGqbCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"854539be95c9ba026d51070b0b13cf3845db4b6d8936f37ca0003c21a4b3eb99","last_reissued_at":"2026-07-28T00:21:53.346118Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T00:21:53.346118Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Shapley: An Influence-Based Data Auditing Pipeline for LLM Alignment and Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jun Qin, Matthew Watson, Peter Grabowski, Yunting Song","submitted_at":"2026-07-24T03:31:44Z","abstract_excerpt":"The alignment of Large Language Models (LLMs) is increasingly bottlenecked by data quality. As datasets scale, massive preference and instruction-tuning corpora inevitably accumulate hidden structural contradictions, safety risks, and systemic human annotation errors. Standard dataset auditing methods, such as semantic deduplication or LLM-as-a-judge, struggle to capture the actual predictive impact of individual records and often miss deep functional rule clashes. To address this, we introduce a scalable, inference-only data valuation pipeline that approximates the Shapley value without itera"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22766","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/2607.22766/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":"2607.22766","created_at":"2026-07-28T00:21:53.346565+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.22766v1","created_at":"2026-07-28T00:21:53.346565+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22766","created_at":"2026-07-28T00:21:53.346565+00:00"},{"alias_kind":"pith_short_12","alias_value":"QVCTTPUVZG5A","created_at":"2026-07-28T00:21:53.346565+00:00"},{"alias_kind":"pith_short_16","alias_value":"QVCTTPUVZG5AE3KR","created_at":"2026-07-28T00:21:53.346565+00:00"},{"alias_kind":"pith_short_8","alias_value":"QVCTTPUV","created_at":"2026-07-28T00:21:53.346565+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/QVCTTPUVZG5AE3KRA4FQWE6PHB","json":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB.json","graph_json":"https://pith.science/api/pith-number/QVCTTPUVZG5AE3KRA4FQWE6PHB/graph.json","events_json":"https://pith.science/api/pith-number/QVCTTPUVZG5AE3KRA4FQWE6PHB/events.json","paper":"https://pith.science/paper/QVCTTPUV"},"agent_actions":{"view_html":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB","download_json":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB.json","view_paper":"https://pith.science/paper/QVCTTPUV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.22766&json=true","fetch_graph":"https://pith.science/api/pith-number/QVCTTPUVZG5AE3KRA4FQWE6PHB/graph.json","fetch_events":"https://pith.science/api/pith-number/QVCTTPUVZG5AE3KRA4FQWE6PHB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB/action/storage_attestation","attest_author":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB/action/author_attestation","sign_citation":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB/action/citation_signature","submit_replication":"https://pith.science/pith/QVCTTPUVZG5AE3KRA4FQWE6PHB/action/replication_record"}},"created_at":"2026-07-28T00:21:53.346565+00:00","updated_at":"2026-07-28T00:21:53.346565+00:00"}