{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:UPA2HFP5LOEGUP5OWQLAZTHHTZ","short_pith_number":"pith:UPA2HFP5","schema_version":"1.0","canonical_sha256":"a3c1a395fd5b886a3faeb4160ccce79e75d3d68a177ca230c4cc00e4e7853c94","source":{"kind":"arxiv","id":"2603.18528","version":2},"attestation_state":"computed","paper":{"title":"Correlation-Weighted Multi-Reward Optimization for Compositional Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Donghyun Kim, Hyunsoo Kim, Jungmyung Wi","submitted_at":"2026-03-19T06:19:18Z","abstract_excerpt":"Text-to-image models produce images that align well with natural language prompts, but compositional generation has long been a central challenge. Models often struggle to satisfy multiple concepts within a single prompt, frequently omitting some concepts and resulting in partial success. Such failures highlight the difficulty of jointly optimizing multiple concepts during reward optimization, where competing concepts can interfere with one another. To address this limitation, we propose Correlation-Weighted Multi-Reward Optimization (CMO), a framework that leverages the correlation structure "},"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":"2603.18528","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-03-19T06:19:18Z","cross_cats_sorted":[],"title_canon_sha256":"981f13c80b17bdc565309be21f2527f3118cf3fcbed9a07d9e0930c4bf91505f","abstract_canon_sha256":"ffb0995cd7a0e1e86c957fa03570a4ef9a92ba13bb9cee026235fc50f3a43b8a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:19:49.746979Z","signature_b64":"cuwQpJDHZ2+ksvDTyJ3M8WGVVH0UOIEfqGdCMtTZ5hCmlyq9/DqO0qCZ35SkUgNCJJmcEKAlvNSCsVXroXXJBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a3c1a395fd5b886a3faeb4160ccce79e75d3d68a177ca230c4cc00e4e7853c94","last_reissued_at":"2026-07-07T02:19:49.746162Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:19:49.746162Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Correlation-Weighted Multi-Reward Optimization for Compositional Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Donghyun Kim, Hyunsoo Kim, Jungmyung Wi","submitted_at":"2026-03-19T06:19:18Z","abstract_excerpt":"Text-to-image models produce images that align well with natural language prompts, but compositional generation has long been a central challenge. Models often struggle to satisfy multiple concepts within a single prompt, frequently omitting some concepts and resulting in partial success. Such failures highlight the difficulty of jointly optimizing multiple concepts during reward optimization, where competing concepts can interfere with one another. To address this limitation, we propose Correlation-Weighted Multi-Reward Optimization (CMO), a framework that leverages the correlation structure "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.18528","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/2603.18528/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":"2603.18528","created_at":"2026-07-07T02:19:49.746279+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.18528v2","created_at":"2026-07-07T02:19:49.746279+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.18528","created_at":"2026-07-07T02:19:49.746279+00:00"},{"alias_kind":"pith_short_12","alias_value":"UPA2HFP5LOEG","created_at":"2026-07-07T02:19:49.746279+00:00"},{"alias_kind":"pith_short_16","alias_value":"UPA2HFP5LOEGUP5O","created_at":"2026-07-07T02:19:49.746279+00:00"},{"alias_kind":"pith_short_8","alias_value":"UPA2HFP5","created_at":"2026-07-07T02:19:49.746279+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/UPA2HFP5LOEGUP5OWQLAZTHHTZ","json":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ.json","graph_json":"https://pith.science/api/pith-number/UPA2HFP5LOEGUP5OWQLAZTHHTZ/graph.json","events_json":"https://pith.science/api/pith-number/UPA2HFP5LOEGUP5OWQLAZTHHTZ/events.json","paper":"https://pith.science/paper/UPA2HFP5"},"agent_actions":{"view_html":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ","download_json":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ.json","view_paper":"https://pith.science/paper/UPA2HFP5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.18528&json=true","fetch_graph":"https://pith.science/api/pith-number/UPA2HFP5LOEGUP5OWQLAZTHHTZ/graph.json","fetch_events":"https://pith.science/api/pith-number/UPA2HFP5LOEGUP5OWQLAZTHHTZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ/action/storage_attestation","attest_author":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ/action/author_attestation","sign_citation":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ/action/citation_signature","submit_replication":"https://pith.science/pith/UPA2HFP5LOEGUP5OWQLAZTHHTZ/action/replication_record"}},"created_at":"2026-07-07T02:19:49.746279+00:00","updated_at":"2026-07-07T02:19:49.746279+00:00"}