{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:3TYMJP6V6BIZW7DASQMWKRTW3H","short_pith_number":"pith:3TYMJP6V","schema_version":"1.0","canonical_sha256":"dcf0c4bfd5f0519b7c609419654676d9cf31a65a2938f26f8f90ee46eb101766","source":{"kind":"arxiv","id":"2409.11378","version":1},"attestation_state":"computed","paper":{"title":"Diversify and Conquer: Diversity-Centric Data Selection with Iterative Refinement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Liangyu Chen, Marzieh Fadaee, Sara Ahmadian, Simon Yu","submitted_at":"2024-09-17T17:25:31Z","abstract_excerpt":"Finetuning large language models on instruction data is crucial for enhancing pre-trained knowledge and improving instruction-following capabilities. As instruction datasets proliferate, selecting optimal data for effective training becomes increasingly important. This work addresses the question: How can we determine the optimal subset of data for effective training? While existing research often emphasizes local criteria like instance quality for subset selection, we argue that a global approach focused on data diversity is more critical. Our method employs k-means clustering to ensure the s"},"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":"2409.11378","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-17T17:25:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"5d20e36be7c28af0bcad074f947824bee814e9cb2dbaa7176266ee4a94929000","abstract_canon_sha256":"6f4e0756030981b74e200bc9acb8d8d30429455fa37eb7ddb9453aa68a134be5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:08:17.265498Z","signature_b64":"yvh/HbvWRLv0vzjV2bgOPwq3P7ynbNIX6fcaqR1nfUJc2wQ0RM+8j+xbpKwQAh77ygxK7exXI0uoRVVLKJCbCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"dcf0c4bfd5f0519b7c609419654676d9cf31a65a2938f26f8f90ee46eb101766","last_reissued_at":"2026-07-05T09:08:17.264996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:08:17.264996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diversify and Conquer: Diversity-Centric Data Selection with Iterative Refinement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Liangyu Chen, Marzieh Fadaee, Sara Ahmadian, Simon Yu","submitted_at":"2024-09-17T17:25:31Z","abstract_excerpt":"Finetuning large language models on instruction data is crucial for enhancing pre-trained knowledge and improving instruction-following capabilities. As instruction datasets proliferate, selecting optimal data for effective training becomes increasingly important. This work addresses the question: How can we determine the optimal subset of data for effective training? While existing research often emphasizes local criteria like instance quality for subset selection, we argue that a global approach focused on data diversity is more critical. Our method employs k-means clustering to ensure the s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.11378","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/2409.11378/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":"2409.11378","created_at":"2026-07-05T09:08:17.265055+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.11378v1","created_at":"2026-07-05T09:08:17.265055+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.11378","created_at":"2026-07-05T09:08:17.265055+00:00"},{"alias_kind":"pith_short_12","alias_value":"3TYMJP6V6BIZ","created_at":"2026-07-05T09:08:17.265055+00:00"},{"alias_kind":"pith_short_16","alias_value":"3TYMJP6V6BIZW7DA","created_at":"2026-07-05T09:08:17.265055+00:00"},{"alias_kind":"pith_short_8","alias_value":"3TYMJP6V","created_at":"2026-07-05T09:08:17.265055+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.11810","citing_title":"GRACE: A Dynamic Coreset Selection Framework for Large Language Model Optimization","ref_index":79,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H","json":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H.json","graph_json":"https://pith.science/api/pith-number/3TYMJP6V6BIZW7DASQMWKRTW3H/graph.json","events_json":"https://pith.science/api/pith-number/3TYMJP6V6BIZW7DASQMWKRTW3H/events.json","paper":"https://pith.science/paper/3TYMJP6V"},"agent_actions":{"view_html":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H","download_json":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H.json","view_paper":"https://pith.science/paper/3TYMJP6V","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.11378&json=true","fetch_graph":"https://pith.science/api/pith-number/3TYMJP6V6BIZW7DASQMWKRTW3H/graph.json","fetch_events":"https://pith.science/api/pith-number/3TYMJP6V6BIZW7DASQMWKRTW3H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H/action/storage_attestation","attest_author":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H/action/author_attestation","sign_citation":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H/action/citation_signature","submit_replication":"https://pith.science/pith/3TYMJP6V6BIZW7DASQMWKRTW3H/action/replication_record"}},"created_at":"2026-07-05T09:08:17.265055+00:00","updated_at":"2026-07-05T09:08:17.265055+00:00"}