{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IND5BG7BSHJKAT5UIFTPXQ5SMI","short_pith_number":"pith:IND5BG7B","schema_version":"1.0","canonical_sha256":"4347d09be191d2a04fb44166fbc3b26211a7178c044cdc75e95112592ea2188e","source":{"kind":"arxiv","id":"2504.13835","version":1},"attestation_state":"computed","paper":{"title":"MIG: Automatic Data Selection for Instruction Tuning by Maximizing Information Gain in Semantic Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haochen Ye, Kai Chen, Kai Hu, Yicheng Chen, Yining Li, Zerun Ma","submitted_at":"2025-04-18T17:59:46Z","abstract_excerpt":"Data quality and diversity are key to the construction of effective instruction-tuning datasets. %\nWith the increasing availability of open-source instruction-tuning datasets, it is advantageous to automatically select high-quality and diverse subsets from a vast amount of data. %\nExisting methods typically prioritize instance quality and use heuristic rules to maintain diversity. %\nHowever, this absence of a comprehensive view of the entire collection often leads to suboptimal results. %\nMoreover, heuristic rules generally focus on distance or clustering within the embedding space, which fail"},"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":"2504.13835","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-18T17:59:46Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"35c3915f6d539e5da9bbf81cb26383b9880bc847a0e60f363ea13150c3609b8d","abstract_canon_sha256":"0f0250c98521ac30cb57bfeacd8634ee6c48fc195e3feb5cef56bda45d899524"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:51:02.760800Z","signature_b64":"vBa5lbR//VGkVpA1PxkaHavYKaFlt840oCtVw8scSPIe2LdhcZFRebN4Fu01GZ5O0CAQonY8PmmODvdJpH3fBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4347d09be191d2a04fb44166fbc3b26211a7178c044cdc75e95112592ea2188e","last_reissued_at":"2026-07-05T10:51:02.760206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:51:02.760206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MIG: Automatic Data Selection for Instruction Tuning by Maximizing Information Gain in Semantic Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Haochen Ye, Kai Chen, Kai Hu, Yicheng Chen, Yining Li, Zerun Ma","submitted_at":"2025-04-18T17:59:46Z","abstract_excerpt":"Data quality and diversity are key to the construction of effective instruction-tuning datasets. %\nWith the increasing availability of open-source instruction-tuning datasets, it is advantageous to automatically select high-quality and diverse subsets from a vast amount of data. %\nExisting methods typically prioritize instance quality and use heuristic rules to maintain diversity. %\nHowever, this absence of a comprehensive view of the entire collection often leads to suboptimal results. %\nMoreover, heuristic rules generally focus on distance or clustering within the embedding space, which fail"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.13835","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/2504.13835/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":"2504.13835","created_at":"2026-07-05T10:51:02.760271+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.13835v1","created_at":"2026-07-05T10:51:02.760271+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.13835","created_at":"2026-07-05T10:51:02.760271+00:00"},{"alias_kind":"pith_short_12","alias_value":"IND5BG7BSHJK","created_at":"2026-07-05T10:51:02.760271+00:00"},{"alias_kind":"pith_short_16","alias_value":"IND5BG7BSHJKAT5U","created_at":"2026-07-05T10:51:02.760271+00:00"},{"alias_kind":"pith_short_8","alias_value":"IND5BG7B","created_at":"2026-07-05T10:51:02.760271+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.17082","citing_title":"GemMaroc: Unlocking Darija Proficiency in LLMs with Minimal Data","ref_index":2024,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI","json":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI.json","graph_json":"https://pith.science/api/pith-number/IND5BG7BSHJKAT5UIFTPXQ5SMI/graph.json","events_json":"https://pith.science/api/pith-number/IND5BG7BSHJKAT5UIFTPXQ5SMI/events.json","paper":"https://pith.science/paper/IND5BG7B"},"agent_actions":{"view_html":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI","download_json":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI.json","view_paper":"https://pith.science/paper/IND5BG7B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.13835&json=true","fetch_graph":"https://pith.science/api/pith-number/IND5BG7BSHJKAT5UIFTPXQ5SMI/graph.json","fetch_events":"https://pith.science/api/pith-number/IND5BG7BSHJKAT5UIFTPXQ5SMI/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI/action/storage_attestation","attest_author":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI/action/author_attestation","sign_citation":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI/action/citation_signature","submit_replication":"https://pith.science/pith/IND5BG7BSHJKAT5UIFTPXQ5SMI/action/replication_record"}},"created_at":"2026-07-05T10:51:02.760271+00:00","updated_at":"2026-07-05T10:51:02.760271+00:00"}