{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AWHLPH6RGDOKUTJGEW4TFFLBMW","short_pith_number":"pith:AWHLPH6R","schema_version":"1.0","canonical_sha256":"058eb79fd130dcaa4d2625b9329561658160cbba010dfa78dad3f7978f42e866","source":{"kind":"arxiv","id":"2410.15475","version":2},"attestation_state":"computed","paper":{"title":"Multimodal Fusion via Self-Consistent Task-Gradient Fields","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jianlong Kwan, Jiayu Xiong, Jing Wang, Jun Xue, Wanlong Wang, Xiaosen Lyu, Zhouqiang Jiang","submitted_at":"2024-10-20T19:15:28Z","abstract_excerpt":"Multimodal learning aims to preserve as much task-related information as possible from different inputs. However, current fusion designs often distort the feedback loop to feature extractors. Aggressively merging modalities entangles their representations, making the feature extractors fragile to incomplete inputs. Meanwhile, attempting to separate features via auxiliary losses frequently introduces optimization conflicts that distract from the primary task. We propose the Self-Consistent Field Autoencoder (SCFAE) to provide a better path for task gradients. Our method follows the self-consist"},"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":"2410.15475","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-10-20T19:15:28Z","cross_cats_sorted":[],"title_canon_sha256":"02def66291d5fff985d6760410f1ce5ad5203c55b84ea293f8880274d700430d","abstract_canon_sha256":"7c59d034eab46c49f26d339944769a6d66638b462564c4fd0a7132a5e8e90f08"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:39.132012Z","signature_b64":"dc6TilVF0HPvprzRAHI3z7WuzBOJQz4G/jodq+MYXlYJlE70JYB8NrsoYys/nz/MooOnLHiG7UeENHS7oSxuAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"058eb79fd130dcaa4d2625b9329561658160cbba010dfa78dad3f7978f42e866","last_reissued_at":"2026-06-01T01:03:39.131062Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:39.131062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Multimodal Fusion via Self-Consistent Task-Gradient Fields","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jianlong Kwan, Jiayu Xiong, Jing Wang, Jun Xue, Wanlong Wang, Xiaosen Lyu, Zhouqiang Jiang","submitted_at":"2024-10-20T19:15:28Z","abstract_excerpt":"Multimodal learning aims to preserve as much task-related information as possible from different inputs. However, current fusion designs often distort the feedback loop to feature extractors. Aggressively merging modalities entangles their representations, making the feature extractors fragile to incomplete inputs. Meanwhile, attempting to separate features via auxiliary losses frequently introduces optimization conflicts that distract from the primary task. We propose the Self-Consistent Field Autoencoder (SCFAE) to provide a better path for task gradients. Our method follows the self-consist"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.15475","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/2410.15475/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":"2410.15475","created_at":"2026-06-01T01:03:39.131225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.15475v2","created_at":"2026-06-01T01:03:39.131225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.15475","created_at":"2026-06-01T01:03:39.131225+00:00"},{"alias_kind":"pith_short_12","alias_value":"AWHLPH6RGDOK","created_at":"2026-06-01T01:03:39.131225+00:00"},{"alias_kind":"pith_short_16","alias_value":"AWHLPH6RGDOKUTJG","created_at":"2026-06-01T01:03:39.131225+00:00"},{"alias_kind":"pith_short_8","alias_value":"AWHLPH6R","created_at":"2026-06-01T01:03:39.131225+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/AWHLPH6RGDOKUTJGEW4TFFLBMW","json":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW.json","graph_json":"https://pith.science/api/pith-number/AWHLPH6RGDOKUTJGEW4TFFLBMW/graph.json","events_json":"https://pith.science/api/pith-number/AWHLPH6RGDOKUTJGEW4TFFLBMW/events.json","paper":"https://pith.science/paper/AWHLPH6R"},"agent_actions":{"view_html":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW","download_json":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW.json","view_paper":"https://pith.science/paper/AWHLPH6R","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.15475&json=true","fetch_graph":"https://pith.science/api/pith-number/AWHLPH6RGDOKUTJGEW4TFFLBMW/graph.json","fetch_events":"https://pith.science/api/pith-number/AWHLPH6RGDOKUTJGEW4TFFLBMW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW/action/storage_attestation","attest_author":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW/action/author_attestation","sign_citation":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW/action/citation_signature","submit_replication":"https://pith.science/pith/AWHLPH6RGDOKUTJGEW4TFFLBMW/action/replication_record"}},"created_at":"2026-06-01T01:03:39.131225+00:00","updated_at":"2026-06-01T01:03:39.131225+00:00"}