{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WYUSTE3DESWSVC4NUOGQOJLV5N","short_pith_number":"pith:WYUSTE3D","schema_version":"1.0","canonical_sha256":"b62929936324ad2a8b8da38d072575eb6286fa8a3cc0fa50bacdae8d010beaf0","source":{"kind":"arxiv","id":"2306.09373","version":2},"attestation_state":"computed","paper":{"title":"Equitable Multi-task Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jun Yuan, Rui Zhang","submitted_at":"2023-06-15T03:37:23Z","abstract_excerpt":"Multi-task learning (MTL) has achieved great success in various research domains, such as CV, NLP and IR etc. Due to the complex and competing task correlation, naive training all tasks may lead to inequitable learning, i.e. some tasks are learned well while others are overlooked. Multi-task optimization (MTO) aims to improve all tasks at same time, but conventional methods often perform poor when tasks with large loss scale or gradient norm magnitude difference. To solve the issue, we in-depth investigate the equity problem for MTL and find that regularizing relative contribution of different"},"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":"2306.09373","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-15T03:37:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8443b60521998d4414efd6f542da7d03c1ac369982cff1b239eddbfbcb19ee7a","abstract_canon_sha256":"679a2ec9edd77e491b506842afb43324df1e911ee08be8cc845d29ed55bdd909"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:22:07.604699Z","signature_b64":"OSjTBZAHt3VDRtbu2J021iU3ZvJ8wj5A76Cou7zPsL9kAY8TZu30syyX+NUrFnaZNlm2oMTOalsWWu005CneDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b62929936324ad2a8b8da38d072575eb6286fa8a3cc0fa50bacdae8d010beaf0","last_reissued_at":"2026-07-05T06:22:07.604293Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:22:07.604293Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Equitable Multi-task Learning","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jun Yuan, Rui Zhang","submitted_at":"2023-06-15T03:37:23Z","abstract_excerpt":"Multi-task learning (MTL) has achieved great success in various research domains, such as CV, NLP and IR etc. Due to the complex and competing task correlation, naive training all tasks may lead to inequitable learning, i.e. some tasks are learned well while others are overlooked. Multi-task optimization (MTO) aims to improve all tasks at same time, but conventional methods often perform poor when tasks with large loss scale or gradient norm magnitude difference. To solve the issue, we in-depth investigate the equity problem for MTL and find that regularizing relative contribution of different"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.09373","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/2306.09373/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":"2306.09373","created_at":"2026-07-05T06:22:07.604360+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.09373v2","created_at":"2026-07-05T06:22:07.604360+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.09373","created_at":"2026-07-05T06:22:07.604360+00:00"},{"alias_kind":"pith_short_12","alias_value":"WYUSTE3DESWS","created_at":"2026-07-05T06:22:07.604360+00:00"},{"alias_kind":"pith_short_16","alias_value":"WYUSTE3DESWSVC4N","created_at":"2026-07-05T06:22:07.604360+00:00"},{"alias_kind":"pith_short_8","alias_value":"WYUSTE3D","created_at":"2026-07-05T06:22:07.604360+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N","json":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N.json","graph_json":"https://pith.science/api/pith-number/WYUSTE3DESWSVC4NUOGQOJLV5N/graph.json","events_json":"https://pith.science/api/pith-number/WYUSTE3DESWSVC4NUOGQOJLV5N/events.json","paper":"https://pith.science/paper/WYUSTE3D"},"agent_actions":{"view_html":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N","download_json":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N.json","view_paper":"https://pith.science/paper/WYUSTE3D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.09373&json=true","fetch_graph":"https://pith.science/api/pith-number/WYUSTE3DESWSVC4NUOGQOJLV5N/graph.json","fetch_events":"https://pith.science/api/pith-number/WYUSTE3DESWSVC4NUOGQOJLV5N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N/action/storage_attestation","attest_author":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N/action/author_attestation","sign_citation":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N/action/citation_signature","submit_replication":"https://pith.science/pith/WYUSTE3DESWSVC4NUOGQOJLV5N/action/replication_record"}},"created_at":"2026-07-05T06:22:07.604360+00:00","updated_at":"2026-07-05T06:22:07.604360+00:00"}