{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:DOFG4J3KTGM5FXKUXLVNNMJR7B","short_pith_number":"pith:DOFG4J3K","schema_version":"1.0","canonical_sha256":"1b8a6e276a9999d2dd54baead6b131f852e09e491d1d5a77d2aa5b0f7f58c3ee","source":{"kind":"arxiv","id":"2405.20192","version":2},"attestation_state":"computed","paper":{"title":"TAIA: Large Language Models are Out-of-Distribution Data Learners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shuyang Jiang, Yanfeng Wang, Ya Zhang, Yusheng Liao, Yu Wang","submitted_at":"2024-05-30T15:57:19Z","abstract_excerpt":"Fine-tuning on task-specific question-answer pairs is a predominant method for enhancing the performance of instruction-tuned large language models (LLMs) on downstream tasks. However, in certain specialized domains, such as healthcare or harmless content generation, it is nearly impossible to obtain a large volume of high-quality data that matches the downstream distribution. To improve the performance of LLMs in data-scarce domains with domain-mismatched data, we re-evaluated the Transformer architecture and discovered that not all parameter updates during fine-tuning contribute positively t"},"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":"2405.20192","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-05-30T15:57:19Z","cross_cats_sorted":[],"title_canon_sha256":"a3f84f6d43a2ed0cc4cb7777e2c675d5b57086fb5d280fed27415c31586f6347","abstract_canon_sha256":"708db0f4b5931ef984388a79cbb2aa42e49e09ee3d71244b77a8098ba5c8743b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:21:44.818414Z","signature_b64":"qZFwRnp9zv/PQWPm2fC2+bqwm6q3ezNxX15VKpI1Qvujgm0kqbqLbbmwLV4OyA9z08joBdXKz5ARaToB4eJADQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b8a6e276a9999d2dd54baead6b131f852e09e491d1d5a77d2aa5b0f7f58c3ee","last_reissued_at":"2026-07-05T09:21:44.817888Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:21:44.817888Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TAIA: Large Language Models are Out-of-Distribution Data Learners","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shuyang Jiang, Yanfeng Wang, Ya Zhang, Yusheng Liao, Yu Wang","submitted_at":"2024-05-30T15:57:19Z","abstract_excerpt":"Fine-tuning on task-specific question-answer pairs is a predominant method for enhancing the performance of instruction-tuned large language models (LLMs) on downstream tasks. However, in certain specialized domains, such as healthcare or harmless content generation, it is nearly impossible to obtain a large volume of high-quality data that matches the downstream distribution. To improve the performance of LLMs in data-scarce domains with domain-mismatched data, we re-evaluated the Transformer architecture and discovered that not all parameter updates during fine-tuning contribute positively t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.20192","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/2405.20192/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":"2405.20192","created_at":"2026-07-05T09:21:44.817949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.20192v2","created_at":"2026-07-05T09:21:44.817949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.20192","created_at":"2026-07-05T09:21:44.817949+00:00"},{"alias_kind":"pith_short_12","alias_value":"DOFG4J3KTGM5","created_at":"2026-07-05T09:21:44.817949+00:00"},{"alias_kind":"pith_short_16","alias_value":"DOFG4J3KTGM5FXKU","created_at":"2026-07-05T09:21:44.817949+00:00"},{"alias_kind":"pith_short_8","alias_value":"DOFG4J3K","created_at":"2026-07-05T09:21:44.817949+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.22370","citing_title":"SplitLoRA: Balancing Stability and Plasticity in Continual Learning Through Gradient Space Splitting","ref_index":10,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B","json":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B.json","graph_json":"https://pith.science/api/pith-number/DOFG4J3KTGM5FXKUXLVNNMJR7B/graph.json","events_json":"https://pith.science/api/pith-number/DOFG4J3KTGM5FXKUXLVNNMJR7B/events.json","paper":"https://pith.science/paper/DOFG4J3K"},"agent_actions":{"view_html":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B","download_json":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B.json","view_paper":"https://pith.science/paper/DOFG4J3K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.20192&json=true","fetch_graph":"https://pith.science/api/pith-number/DOFG4J3KTGM5FXKUXLVNNMJR7B/graph.json","fetch_events":"https://pith.science/api/pith-number/DOFG4J3KTGM5FXKUXLVNNMJR7B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B/action/storage_attestation","attest_author":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B/action/author_attestation","sign_citation":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B/action/citation_signature","submit_replication":"https://pith.science/pith/DOFG4J3KTGM5FXKUXLVNNMJR7B/action/replication_record"}},"created_at":"2026-07-05T09:21:44.817949+00:00","updated_at":"2026-07-05T09:21:44.817949+00:00"}