{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RTFNTMVHBHR6GEYACF2LXTB23E","short_pith_number":"pith:RTFNTMVH","schema_version":"1.0","canonical_sha256":"8ccad9b2a709e3e313001174bbcc3ad90416be157a50dfe2817230b8ca597f6e","source":{"kind":"arxiv","id":"2411.18615","version":1},"attestation_state":"computed","paper":{"title":"Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Congfeng Cao, Ekaterina Shutova, Gaole Dai, Jiayi Shen, Qizhe Zhang, Shanghang Zhang, Shiji Zhou, Zhi Zhang","submitted_at":"2024-11-27T18:58:22Z","abstract_excerpt":"Advancing towards generalist agents necessitates the concurrent processing of multiple tasks using a unified model, thereby underscoring the growing significance of simultaneous model training on multiple downstream tasks. A common issue in multi-task learning is the occurrence of gradient conflict, which leads to potential competition among different tasks during joint training. This competition often results in improvements in one task at the expense of deterioration in another. Although several optimization methods have been developed to address this issue by manipulating task gradients for"},"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":"2411.18615","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-11-27T18:58:22Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"cf46dd42d6b29c183198744a6947b13bdf171282f54ad9366d73c1a0ea02cc1f","abstract_canon_sha256":"49aebb60f75df90a18c9188632f4979913293b779940d90e786229361d7bae1b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:41:24.156088Z","signature_b64":"1auKEDYo6QxGK0cjFvqH5qCRsEeyDIvb8qjILUVHE3Ki02QiDkDtKttVRXHN3TubLX4l+Yw3upiOAC1AZLm6Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8ccad9b2a709e3e313001174bbcc3ad90416be157a50dfe2817230b8ca597f6e","last_reissued_at":"2026-07-05T09:41:24.155528Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:41:24.155528Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.LG","authors_text":"Congfeng Cao, Ekaterina Shutova, Gaole Dai, Jiayi Shen, Qizhe Zhang, Shanghang Zhang, Shiji Zhou, Zhi Zhang","submitted_at":"2024-11-27T18:58:22Z","abstract_excerpt":"Advancing towards generalist agents necessitates the concurrent processing of multiple tasks using a unified model, thereby underscoring the growing significance of simultaneous model training on multiple downstream tasks. A common issue in multi-task learning is the occurrence of gradient conflict, which leads to potential competition among different tasks during joint training. This competition often results in improvements in one task at the expense of deterioration in another. Although several optimization methods have been developed to address this issue by manipulating task gradients for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.18615","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/2411.18615/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":"2411.18615","created_at":"2026-07-05T09:41:24.155596+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.18615v1","created_at":"2026-07-05T09:41:24.155596+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.18615","created_at":"2026-07-05T09:41:24.155596+00:00"},{"alias_kind":"pith_short_12","alias_value":"RTFNTMVHBHR6","created_at":"2026-07-05T09:41:24.155596+00:00"},{"alias_kind":"pith_short_16","alias_value":"RTFNTMVHBHR6GEYA","created_at":"2026-07-05T09:41:24.155596+00:00"},{"alias_kind":"pith_short_8","alias_value":"RTFNTMVH","created_at":"2026-07-05T09:41:24.155596+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.13560","citing_title":"Parameter-efficient Quantum Multi-task Learning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17257","citing_title":"REZE: Representation Regularization for Domain-adaptive Text Embedding Pre-finetuning","ref_index":70,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E","json":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E.json","graph_json":"https://pith.science/api/pith-number/RTFNTMVHBHR6GEYACF2LXTB23E/graph.json","events_json":"https://pith.science/api/pith-number/RTFNTMVHBHR6GEYACF2LXTB23E/events.json","paper":"https://pith.science/paper/RTFNTMVH"},"agent_actions":{"view_html":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E","download_json":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E.json","view_paper":"https://pith.science/paper/RTFNTMVH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.18615&json=true","fetch_graph":"https://pith.science/api/pith-number/RTFNTMVHBHR6GEYACF2LXTB23E/graph.json","fetch_events":"https://pith.science/api/pith-number/RTFNTMVHBHR6GEYACF2LXTB23E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E/action/storage_attestation","attest_author":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E/action/author_attestation","sign_citation":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E/action/citation_signature","submit_replication":"https://pith.science/pith/RTFNTMVHBHR6GEYACF2LXTB23E/action/replication_record"}},"created_at":"2026-07-05T09:41:24.155596+00:00","updated_at":"2026-07-05T09:41:24.155596+00:00"}