{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:QD64J2SYZ626NJ5RL6KHBAFXH4","short_pith_number":"pith:QD64J2SY","schema_version":"1.0","canonical_sha256":"80fdc4ea58cfb5e6a7b15f947080b73f10f402e17f2391b2f44154545bf899d7","source":{"kind":"arxiv","id":"2406.19976","version":2},"attestation_state":"computed","paper":{"title":"ScaleBiO: Scalable Bilevel Optimization for LLM Data Reweighting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Dylan Zhang, Hanning Zhang, Jipeng Zhang, Minrui Xu, Renjie Pi, Rui Pan, Tong Zhang, Xiaoyu Wang, Xingyuan Pan","submitted_at":"2024-06-28T15:03:08Z","abstract_excerpt":"Bilevel optimization has shown its utility across various machine learning settings, yet most algorithms in practice require second-order information, making it challenging to scale them up. Only recently, a paradigm of first-order algorithms has emerged in the theoretical literature, capable of effectively addressing bilevel optimization problems. Nevertheless, the practical efficiency of this paradigm remains unverified, particularly in the context of large language models (LLMs). This paper introduces the first scalable instantiation of this paradigm called ScaleBiO, focusing on bilevel opt"},"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":"2406.19976","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-06-28T15:03:08Z","cross_cats_sorted":["math.OC"],"title_canon_sha256":"947b46be9120ac1115d842ece17a894181e33dd6390ec8ad81bc899ff849de7e","abstract_canon_sha256":"b577d675293dcb5c493da94fe210ab5b260bf62314d161d3d4b3307379e042e9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:08:58.575971Z","signature_b64":"exhrfWFrSBiSHgbnLC3iKqZkbMz7GTgZ+sfZqJuXIcRaNbBaT1RuyEi0feA9ajDCkAkV8m0LZbKl8VDeM0DYBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"80fdc4ea58cfb5e6a7b15f947080b73f10f402e17f2391b2f44154545bf899d7","last_reissued_at":"2026-07-05T11:08:58.575401Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:08:58.575401Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ScaleBiO: Scalable Bilevel Optimization for LLM Data Reweighting","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["math.OC"],"primary_cat":"cs.LG","authors_text":"Dylan Zhang, Hanning Zhang, Jipeng Zhang, Minrui Xu, Renjie Pi, Rui Pan, Tong Zhang, Xiaoyu Wang, Xingyuan Pan","submitted_at":"2024-06-28T15:03:08Z","abstract_excerpt":"Bilevel optimization has shown its utility across various machine learning settings, yet most algorithms in practice require second-order information, making it challenging to scale them up. Only recently, a paradigm of first-order algorithms has emerged in the theoretical literature, capable of effectively addressing bilevel optimization problems. Nevertheless, the practical efficiency of this paradigm remains unverified, particularly in the context of large language models (LLMs). This paper introduces the first scalable instantiation of this paradigm called ScaleBiO, focusing on bilevel opt"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.19976","kind":"arxiv","version":2},"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/2406.19976/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":"2406.19976","created_at":"2026-07-05T11:08:58.575485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.19976v2","created_at":"2026-07-05T11:08:58.575485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.19976","created_at":"2026-07-05T11:08:58.575485+00:00"},{"alias_kind":"pith_short_12","alias_value":"QD64J2SYZ626","created_at":"2026-07-05T11:08:58.575485+00:00"},{"alias_kind":"pith_short_16","alias_value":"QD64J2SYZ626NJ5R","created_at":"2026-07-05T11:08:58.575485+00:00"},{"alias_kind":"pith_short_8","alias_value":"QD64J2SY","created_at":"2026-07-05T11:08:58.575485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04401","citing_title":"TANDEM: Bi-Level Data Mixture Optimization with Twin Networks","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14350","citing_title":"Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling","ref_index":242,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4","json":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4.json","graph_json":"https://pith.science/api/pith-number/QD64J2SYZ626NJ5RL6KHBAFXH4/graph.json","events_json":"https://pith.science/api/pith-number/QD64J2SYZ626NJ5RL6KHBAFXH4/events.json","paper":"https://pith.science/paper/QD64J2SY"},"agent_actions":{"view_html":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4","download_json":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4.json","view_paper":"https://pith.science/paper/QD64J2SY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.19976&json=true","fetch_graph":"https://pith.science/api/pith-number/QD64J2SYZ626NJ5RL6KHBAFXH4/graph.json","fetch_events":"https://pith.science/api/pith-number/QD64J2SYZ626NJ5RL6KHBAFXH4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4/action/storage_attestation","attest_author":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4/action/author_attestation","sign_citation":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4/action/citation_signature","submit_replication":"https://pith.science/pith/QD64J2SYZ626NJ5RL6KHBAFXH4/action/replication_record"}},"created_at":"2026-07-05T11:08:58.575485+00:00","updated_at":"2026-07-05T11:08:58.575485+00:00"}