{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LLNNHA5ZLJ2WRMKVP5NM6537V3","short_pith_number":"pith:LLNNHA5Z","schema_version":"1.0","canonical_sha256":"5adad383b95a7568b1557f5acf777faee37c89322f97d4153874cb65d5153a28","source":{"kind":"arxiv","id":"2605.30537","version":1},"attestation_state":"computed","paper":{"title":"The Long-Term Effects of Data Selection in LLM Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aoxiong Zeng, Xiangquan Yang, Yuxin Yang","submitted_at":"2026-05-28T20:12:45Z","abstract_excerpt":"Data selection is increasingly used to reduce the cost of large language model (LLM) fine-tuning, with recent methods prioritizing samples by current utility, diversity, quality, or influence. This paper studies a different question: when fine-tuning occurs over multiple stages, can selection strategies that look optimal now make the model less adaptable later? We introduce a long-horizon view of LLM data selection in which a selector is evaluated not only by immediate task performance, but also by future adaptation speed, forgetting, capability imbalance, and out-of-distribution robustness. W"},"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":"2605.30537","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-28T20:12:45Z","cross_cats_sorted":[],"title_canon_sha256":"bfaf1fbb882daa65d7db49e656c62e2d2155795cae9db6370646cfe30ef985d2","abstract_canon_sha256":"94cd3262335a86cda3f4be9e6fee1f85992cabcafeafac47bef3348d29be6269"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:59.663634Z","signature_b64":"nK4Ud7B583WYXcSyFvMwtn5IfT3ad+ijdAkbmRwKWUjwZTMXeMpK9sp0cvlIQlAtecaxypZzlVDuWnIabb9RBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5adad383b95a7568b1557f5acf777faee37c89322f97d4153874cb65d5153a28","last_reissued_at":"2026-06-01T01:02:59.662742Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:59.662742Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Long-Term Effects of Data Selection in LLM Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Aoxiong Zeng, Xiangquan Yang, Yuxin Yang","submitted_at":"2026-05-28T20:12:45Z","abstract_excerpt":"Data selection is increasingly used to reduce the cost of large language model (LLM) fine-tuning, with recent methods prioritizing samples by current utility, diversity, quality, or influence. This paper studies a different question: when fine-tuning occurs over multiple stages, can selection strategies that look optimal now make the model less adaptable later? We introduce a long-horizon view of LLM data selection in which a selector is evaluated not only by immediate task performance, but also by future adaptation speed, forgetting, capability imbalance, and out-of-distribution robustness. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.30537","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/2605.30537/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":"2605.30537","created_at":"2026-06-01T01:02:59.662891+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.30537v1","created_at":"2026-06-01T01:02:59.662891+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.30537","created_at":"2026-06-01T01:02:59.662891+00:00"},{"alias_kind":"pith_short_12","alias_value":"LLNNHA5ZLJ2W","created_at":"2026-06-01T01:02:59.662891+00:00"},{"alias_kind":"pith_short_16","alias_value":"LLNNHA5ZLJ2WRMKV","created_at":"2026-06-01T01:02:59.662891+00:00"},{"alias_kind":"pith_short_8","alias_value":"LLNNHA5Z","created_at":"2026-06-01T01:02:59.662891+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/LLNNHA5ZLJ2WRMKVP5NM6537V3","json":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3.json","graph_json":"https://pith.science/api/pith-number/LLNNHA5ZLJ2WRMKVP5NM6537V3/graph.json","events_json":"https://pith.science/api/pith-number/LLNNHA5ZLJ2WRMKVP5NM6537V3/events.json","paper":"https://pith.science/paper/LLNNHA5Z"},"agent_actions":{"view_html":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3","download_json":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3.json","view_paper":"https://pith.science/paper/LLNNHA5Z","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.30537&json=true","fetch_graph":"https://pith.science/api/pith-number/LLNNHA5ZLJ2WRMKVP5NM6537V3/graph.json","fetch_events":"https://pith.science/api/pith-number/LLNNHA5ZLJ2WRMKVP5NM6537V3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3/action/storage_attestation","attest_author":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3/action/author_attestation","sign_citation":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3/action/citation_signature","submit_replication":"https://pith.science/pith/LLNNHA5ZLJ2WRMKVP5NM6537V3/action/replication_record"}},"created_at":"2026-06-01T01:02:59.662891+00:00","updated_at":"2026-06-01T01:02:59.662891+00:00"}