{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:DDHG4QDGCBCI5SQHJCXR2M67F4","short_pith_number":"pith:DDHG4QDG","schema_version":"1.0","canonical_sha256":"18ce6e406610448eca0748af1d33df2f1afae02ac0d223b43e3bf45eea794c4c","source":{"kind":"arxiv","id":"2203.06801","version":1},"attestation_state":"computed","paper":{"title":"MetaBalance: Improving Multi-Task Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.LG","authors_text":"Cheng cheng, Geng Ji, James Caverlee, Xue Feng, Yun He, Yunsong Guo","submitted_at":"2022-03-14T01:08:31Z","abstract_excerpt":"In many personalized recommendation scenarios, the generalization ability of a target task can be improved via learning with additional auxiliary tasks alongside this target task on a multi-task network. However, this method often suffers from a serious optimization imbalance problem. On the one hand, one or more auxiliary tasks might have a larger influence than the target task and even dominate the network weights, resulting in worse recommendation accuracy for the target task. On the other hand, the influence of one or more auxiliary tasks might be too weak to assist the target task. More c"},"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":"2203.06801","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-14T01:08:31Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"b705566c2dc6b3def0e8e3a387e5750e962eb9f67c3f19f649b17ddc774c1fcc","abstract_canon_sha256":"9e0672af026983cf2f976810e5d1b01d3e69ef9118e6941af6e6a0a41f57d9e2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:55.719000Z","signature_b64":"NQbCunF0aR6CyqZ2cH0Zx+1521mrpP38fhBX5grrazm+jM5xFjldTR6kfUWGKuS3jSv2ir7hfCIH4pJX8fl6CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18ce6e406610448eca0748af1d33df2f1afae02ac0d223b43e3bf45eea794c4c","last_reissued_at":"2026-07-05T04:04:55.718457Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:55.718457Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MetaBalance: Improving Multi-Task Recommendations via Adapting Gradient Magnitudes of Auxiliary Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.LG","authors_text":"Cheng cheng, Geng Ji, James Caverlee, Xue Feng, Yun He, Yunsong Guo","submitted_at":"2022-03-14T01:08:31Z","abstract_excerpt":"In many personalized recommendation scenarios, the generalization ability of a target task can be improved via learning with additional auxiliary tasks alongside this target task on a multi-task network. However, this method often suffers from a serious optimization imbalance problem. On the one hand, one or more auxiliary tasks might have a larger influence than the target task and even dominate the network weights, resulting in worse recommendation accuracy for the target task. On the other hand, the influence of one or more auxiliary tasks might be too weak to assist the target task. More c"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.06801","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/2203.06801/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":"2203.06801","created_at":"2026-07-05T04:04:55.718522+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.06801v1","created_at":"2026-07-05T04:04:55.718522+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.06801","created_at":"2026-07-05T04:04:55.718522+00:00"},{"alias_kind":"pith_short_12","alias_value":"DDHG4QDGCBCI","created_at":"2026-07-05T04:04:55.718522+00:00"},{"alias_kind":"pith_short_16","alias_value":"DDHG4QDGCBCI5SQH","created_at":"2026-07-05T04:04:55.718522+00:00"},{"alias_kind":"pith_short_8","alias_value":"DDHG4QDG","created_at":"2026-07-05T04:04:55.718522+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/DDHG4QDGCBCI5SQHJCXR2M67F4","json":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4.json","graph_json":"https://pith.science/api/pith-number/DDHG4QDGCBCI5SQHJCXR2M67F4/graph.json","events_json":"https://pith.science/api/pith-number/DDHG4QDGCBCI5SQHJCXR2M67F4/events.json","paper":"https://pith.science/paper/DDHG4QDG"},"agent_actions":{"view_html":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4","download_json":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4.json","view_paper":"https://pith.science/paper/DDHG4QDG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.06801&json=true","fetch_graph":"https://pith.science/api/pith-number/DDHG4QDGCBCI5SQHJCXR2M67F4/graph.json","fetch_events":"https://pith.science/api/pith-number/DDHG4QDGCBCI5SQHJCXR2M67F4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4/action/storage_attestation","attest_author":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4/action/author_attestation","sign_citation":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4/action/citation_signature","submit_replication":"https://pith.science/pith/DDHG4QDGCBCI5SQHJCXR2M67F4/action/replication_record"}},"created_at":"2026-07-05T04:04:55.718522+00:00","updated_at":"2026-07-05T04:04:55.718522+00:00"}