{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:U4LKF5NWGTDCEQJKBAEFH25ECK","short_pith_number":"pith:U4LKF5NW","schema_version":"1.0","canonical_sha256":"a716a2f5b634c622412a080853eba4129ecf046b42f8bcdbf44aebacf01f0ecb","source":{"kind":"arxiv","id":"2408.09849","version":2},"attestation_state":"computed","paper":{"title":"Importance Weighting Can Help Large Language Models Self-Improve","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chi-Min Chan, Chunyang Jiang, Qifeng Liu, Wei Xue, Yike Guo","submitted_at":"2024-08-19T09:51:02Z","abstract_excerpt":"Large language models (LLMs) have shown remarkable capability in numerous tasks and applications. However, fine-tuning LLMs using high-quality datasets under external supervision remains prohibitively expensive. In response, LLM self-improvement approaches have been vibrantly developed recently. The typical paradigm of LLM self-improvement involves training LLM on self-generated data, part of which may be detrimental and should be filtered out due to the unstable data quality. While current works primarily employs filtering strategies based on answer correctness, in this paper, we demonstrate "},"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":"2408.09849","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-08-19T09:51:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2672097a3ad14f1ebfa0cd0cd93b9f347d575780971809420239dfc99c6e650b","abstract_canon_sha256":"ca6dac792f1963902cefcabaa204b1674fbc2976740087609b28dcd23c2bfbe3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:15.253526Z","signature_b64":"2vlVnwnNxbi2JG+2IgyeBb8hCW6yKfujJ70eRXRZsL/Y9XlkYIwW0ACFkdZ2nLaPHyz3oft1hUcNxvuXoxFyCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a716a2f5b634c622412a080853eba4129ecf046b42f8bcdbf44aebacf01f0ecb","last_reissued_at":"2026-07-05T09:48:15.253048Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:15.253048Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Importance Weighting Can Help Large Language Models Self-Improve","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chi-Min Chan, Chunyang Jiang, Qifeng Liu, Wei Xue, Yike Guo","submitted_at":"2024-08-19T09:51:02Z","abstract_excerpt":"Large language models (LLMs) have shown remarkable capability in numerous tasks and applications. However, fine-tuning LLMs using high-quality datasets under external supervision remains prohibitively expensive. In response, LLM self-improvement approaches have been vibrantly developed recently. The typical paradigm of LLM self-improvement involves training LLM on self-generated data, part of which may be detrimental and should be filtered out due to the unstable data quality. While current works primarily employs filtering strategies based on answer correctness, in this paper, we demonstrate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.09849","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/2408.09849/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":"2408.09849","created_at":"2026-07-05T09:48:15.253104+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.09849v2","created_at":"2026-07-05T09:48:15.253104+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.09849","created_at":"2026-07-05T09:48:15.253104+00:00"},{"alias_kind":"pith_short_12","alias_value":"U4LKF5NWGTDC","created_at":"2026-07-05T09:48:15.253104+00:00"},{"alias_kind":"pith_short_16","alias_value":"U4LKF5NWGTDCEQJK","created_at":"2026-07-05T09:48:15.253104+00:00"},{"alias_kind":"pith_short_8","alias_value":"U4LKF5NW","created_at":"2026-07-05T09:48:15.253104+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.05227","citing_title":"Rethinking Data Curation in LLM Training: Online Reweighting Offers Better Generalization than Offline Methods","ref_index":17,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK","json":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK.json","graph_json":"https://pith.science/api/pith-number/U4LKF5NWGTDCEQJKBAEFH25ECK/graph.json","events_json":"https://pith.science/api/pith-number/U4LKF5NWGTDCEQJKBAEFH25ECK/events.json","paper":"https://pith.science/paper/U4LKF5NW"},"agent_actions":{"view_html":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK","download_json":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK.json","view_paper":"https://pith.science/paper/U4LKF5NW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.09849&json=true","fetch_graph":"https://pith.science/api/pith-number/U4LKF5NWGTDCEQJKBAEFH25ECK/graph.json","fetch_events":"https://pith.science/api/pith-number/U4LKF5NWGTDCEQJKBAEFH25ECK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK/action/storage_attestation","attest_author":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK/action/author_attestation","sign_citation":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK/action/citation_signature","submit_replication":"https://pith.science/pith/U4LKF5NWGTDCEQJKBAEFH25ECK/action/replication_record"}},"created_at":"2026-07-05T09:48:15.253104+00:00","updated_at":"2026-07-05T09:48:15.253104+00:00"}