{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CRSWDVUOMGDWWBWE6PKU5V2CSW","short_pith_number":"pith:CRSWDVUO","schema_version":"1.0","canonical_sha256":"146561d68e61876b06c4f3d54ed74295868ce1013c9eb9a935746ed0789dc65d","source":{"kind":"arxiv","id":"2605.16704","version":1},"attestation_state":"computed","paper":{"title":"Convex Dataset Valuation for Post-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Christopher Jung, Fuchun Peng, Han Zhao, Ming Li, Nima Noorshams, Rui Li, Siqi Zeng, Xue Feng, Zhe Kang, Zhigang Wang","submitted_at":"2026-05-15T23:35:07Z","abstract_excerpt":"Improving LLM performance on downstream tasks sometimes requires leveraging auxiliary datasets during post-training. In practice, however, developers face constraints on compute, labeling, and licensing costs that preclude using all available data, necessitating principled dataset-level selection. These constraints are increasingly shaped by dataset marketplaces, where data acquisition is governed by budgets and negotiation. We study dataset valuation as a subset selection problem during LLM post-training. Our goal is to identify and weight auxiliary datasets so as to maximize target task perf"},"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":"2605.16704","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-15T23:35:07Z","cross_cats_sorted":[],"title_canon_sha256":"4cbebfa04e299b24dce42dccaef6def82411b8ee09e7364998ebaaf768e81625","abstract_canon_sha256":"bb07a10933d5583a7d0b105db060289387572326f41cfdb7a47490b0a4d5aa94"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:37.429077Z","signature_b64":"xeOeJEHLjA5ElkTeKidQIQiz9VwpCIzxPkWdecGQvRH/OTYXh0GBWNmrGiyWZtx1tCvGDm1sigHp52RzsMQeBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"146561d68e61876b06c4f3d54ed74295868ce1013c9eb9a935746ed0789dc65d","last_reissued_at":"2026-05-20T00:02:37.428310Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:37.428310Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Convex Dataset Valuation for Post-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Christopher Jung, Fuchun Peng, Han Zhao, Ming Li, Nima Noorshams, Rui Li, Siqi Zeng, Xue Feng, Zhe Kang, Zhigang Wang","submitted_at":"2026-05-15T23:35:07Z","abstract_excerpt":"Improving LLM performance on downstream tasks sometimes requires leveraging auxiliary datasets during post-training. In practice, however, developers face constraints on compute, labeling, and licensing costs that preclude using all available data, necessitating principled dataset-level selection. These constraints are increasingly shaped by dataset marketplaces, where data acquisition is governed by budgets and negotiation. We study dataset valuation as a subset selection problem during LLM post-training. Our goal is to identify and weight auxiliary datasets so as to maximize target task perf"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.16704","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/2605.16704/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.364449Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.486460Z","status":"skipped","version":"1.0.0","findings_count":0}],"snapshot_sha256":"3cc73d7486932bebf6c331c0bf1b311d5e04c417a6f779deec663f2cd6f48fb1"},"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":"2605.16704","created_at":"2026-05-20T00:02:37.428435+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.16704v1","created_at":"2026-05-20T00:02:37.428435+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16704","created_at":"2026-05-20T00:02:37.428435+00:00"},{"alias_kind":"pith_short_12","alias_value":"CRSWDVUOMGDW","created_at":"2026-05-20T00:02:37.428435+00:00"},{"alias_kind":"pith_short_16","alias_value":"CRSWDVUOMGDWWBWE","created_at":"2026-05-20T00:02:37.428435+00:00"},{"alias_kind":"pith_short_8","alias_value":"CRSWDVUO","created_at":"2026-05-20T00:02:37.428435+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/CRSWDVUOMGDWWBWE6PKU5V2CSW","json":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW.json","graph_json":"https://pith.science/api/pith-number/CRSWDVUOMGDWWBWE6PKU5V2CSW/graph.json","events_json":"https://pith.science/api/pith-number/CRSWDVUOMGDWWBWE6PKU5V2CSW/events.json","paper":"https://pith.science/paper/CRSWDVUO"},"agent_actions":{"view_html":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW","download_json":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW.json","view_paper":"https://pith.science/paper/CRSWDVUO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.16704&json=true","fetch_graph":"https://pith.science/api/pith-number/CRSWDVUOMGDWWBWE6PKU5V2CSW/graph.json","fetch_events":"https://pith.science/api/pith-number/CRSWDVUOMGDWWBWE6PKU5V2CSW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW/action/storage_attestation","attest_author":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW/action/author_attestation","sign_citation":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW/action/citation_signature","submit_replication":"https://pith.science/pith/CRSWDVUOMGDWWBWE6PKU5V2CSW/action/replication_record"}},"created_at":"2026-05-20T00:02:37.428435+00:00","updated_at":"2026-05-20T00:02:37.428435+00:00"}