{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:NWCIDJ73ENT6XHKJVZNMBXWLOY","short_pith_number":"pith:NWCIDJ73","schema_version":"1.0","canonical_sha256":"6d8481a7fb2367eb9d49ae5ac0decb760b2537d16de07bf0fd5a3ac8e8804407","source":{"kind":"arxiv","id":"2508.13979","version":1},"attestation_state":"computed","paper":{"title":"AutoScale: Linear Scalarization Guided by Multi-Task Optimization Metrics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ci Li, John Folkesson, Kei Ikemura, Nazre Batool, Qingwen Zhang, Sina Sharif Mansouri, Xiaomeng Zhu, Yi Yang","submitted_at":"2025-08-19T16:14:00Z","abstract_excerpt":"Recent multi-task learning studies suggest that linear scalarization, when using well-chosen fixed task weights, can achieve comparable to or even better performance than complex multi-task optimization (MTO) methods. It remains unclear why certain weights yield optimal performance and how to determine these weights without relying on exhaustive hyperparameter search. This paper establishes a direct connection between linear scalarization and MTO methods, revealing through extensive experiments that well-performing scalarization weights exhibit specific trends in key MTO metrics, such as high "},"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":"2508.13979","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-08-19T16:14:00Z","cross_cats_sorted":[],"title_canon_sha256":"e2a6968ddfae0d2bb82bff6b9c3ffcabfd6b5db419ab02ad36251b96aca36fb7","abstract_canon_sha256":"08971f9fbbafa99e4a35845fe079af2418a595505bfa56ca17e92f6757320313"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:56:10.359790Z","signature_b64":"LleTb8CNZ7TB43e0w8VZAvfpO74oGWk8Il5r6XJd4ifCe0JK3VAloRL7qwG3d/xWkRoXbJIKi2Snn1h/fic0Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d8481a7fb2367eb9d49ae5ac0decb760b2537d16de07bf0fd5a3ac8e8804407","last_reissued_at":"2026-07-05T11:56:10.359330Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:56:10.359330Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AutoScale: Linear Scalarization Guided by Multi-Task Optimization Metrics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Ci Li, John Folkesson, Kei Ikemura, Nazre Batool, Qingwen Zhang, Sina Sharif Mansouri, Xiaomeng Zhu, Yi Yang","submitted_at":"2025-08-19T16:14:00Z","abstract_excerpt":"Recent multi-task learning studies suggest that linear scalarization, when using well-chosen fixed task weights, can achieve comparable to or even better performance than complex multi-task optimization (MTO) methods. It remains unclear why certain weights yield optimal performance and how to determine these weights without relying on exhaustive hyperparameter search. This paper establishes a direct connection between linear scalarization and MTO methods, revealing through extensive experiments that well-performing scalarization weights exhibit specific trends in key MTO metrics, such as high "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.13979","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/2508.13979/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":"2508.13979","created_at":"2026-07-05T11:56:10.359392+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.13979v1","created_at":"2026-07-05T11:56:10.359392+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.13979","created_at":"2026-07-05T11:56:10.359392+00:00"},{"alias_kind":"pith_short_12","alias_value":"NWCIDJ73ENT6","created_at":"2026-07-05T11:56:10.359392+00:00"},{"alias_kind":"pith_short_16","alias_value":"NWCIDJ73ENT6XHKJ","created_at":"2026-07-05T11:56:10.359392+00:00"},{"alias_kind":"pith_short_8","alias_value":"NWCIDJ73","created_at":"2026-07-05T11:56:10.359392+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/NWCIDJ73ENT6XHKJVZNMBXWLOY","json":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY.json","graph_json":"https://pith.science/api/pith-number/NWCIDJ73ENT6XHKJVZNMBXWLOY/graph.json","events_json":"https://pith.science/api/pith-number/NWCIDJ73ENT6XHKJVZNMBXWLOY/events.json","paper":"https://pith.science/paper/NWCIDJ73"},"agent_actions":{"view_html":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY","download_json":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY.json","view_paper":"https://pith.science/paper/NWCIDJ73","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.13979&json=true","fetch_graph":"https://pith.science/api/pith-number/NWCIDJ73ENT6XHKJVZNMBXWLOY/graph.json","fetch_events":"https://pith.science/api/pith-number/NWCIDJ73ENT6XHKJVZNMBXWLOY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY/action/storage_attestation","attest_author":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY/action/author_attestation","sign_citation":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY/action/citation_signature","submit_replication":"https://pith.science/pith/NWCIDJ73ENT6XHKJVZNMBXWLOY/action/replication_record"}},"created_at":"2026-07-05T11:56:10.359392+00:00","updated_at":"2026-07-05T11:56:10.359392+00:00"}