{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HPYUXF6IAR46MJG4AF53AW5CIA","short_pith_number":"pith:HPYUXF6I","schema_version":"1.0","canonical_sha256":"3bf14b97c80479e624dc017bb05ba240088550234fa67d526099385fbd556ba2","source":{"kind":"arxiv","id":"2406.01210","version":2},"attestation_state":"computed","paper":{"title":"GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chang Xu, Chao Zhang, Ding Jia, Han Wu, Jianyuan Guo, Kai Han, Xinghao Chen","submitted_at":"2024-06-03T11:24:15Z","abstract_excerpt":"Cross-modal transformers have demonstrated superiority in various vision tasks by effectively integrating different modalities. This paper first critiques prior token exchange methods which replace less informative tokens with inter-modal features, and demonstrate exchange based methods underperform cross-attention mechanisms, while the computational demand of the latter inevitably restricts its use with longer sequences. To surmount the computational challenges, we propose GeminiFusion, a pixel-wise fusion approach that capitalizes on aligned cross-modal representations. GeminiFusion elegantl"},"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":"2406.01210","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-03T11:24:15Z","cross_cats_sorted":[],"title_canon_sha256":"f30c35499050b74f14df36218ef2e096957f96b4b22b1c38b642e834dd5bf097","abstract_canon_sha256":"750e77b05540ccc510b4317222c3349b8a97708288dc539589fac8a0f9feb861"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:27:10.798606Z","signature_b64":"eUXBrq/FbmUbuum/Zc58XRfuMj5gtKrKl6mNVm46fH2dNDZksRK729BV6262GuBETIOlX/PB3YbbCIn46hQCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3bf14b97c80479e624dc017bb05ba240088550234fa67d526099385fbd556ba2","last_reissued_at":"2026-07-05T08:27:10.798182Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:27:10.798182Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"GeminiFusion: Efficient Pixel-wise Multimodal Fusion for Vision Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chang Xu, Chao Zhang, Ding Jia, Han Wu, Jianyuan Guo, Kai Han, Xinghao Chen","submitted_at":"2024-06-03T11:24:15Z","abstract_excerpt":"Cross-modal transformers have demonstrated superiority in various vision tasks by effectively integrating different modalities. This paper first critiques prior token exchange methods which replace less informative tokens with inter-modal features, and demonstrate exchange based methods underperform cross-attention mechanisms, while the computational demand of the latter inevitably restricts its use with longer sequences. To surmount the computational challenges, we propose GeminiFusion, a pixel-wise fusion approach that capitalizes on aligned cross-modal representations. GeminiFusion elegantl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.01210","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/2406.01210/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":"2406.01210","created_at":"2026-07-05T08:27:10.798237+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.01210v2","created_at":"2026-07-05T08:27:10.798237+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.01210","created_at":"2026-07-05T08:27:10.798237+00:00"},{"alias_kind":"pith_short_12","alias_value":"HPYUXF6IAR46","created_at":"2026-07-05T08:27:10.798237+00:00"},{"alias_kind":"pith_short_16","alias_value":"HPYUXF6IAR46MJG4","created_at":"2026-07-05T08:27:10.798237+00:00"},{"alias_kind":"pith_short_8","alias_value":"HPYUXF6I","created_at":"2026-07-05T08:27:10.798237+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.02948","citing_title":"CrossWeaver: Cross-modal Weaving for Arbitrary-Modality Semantic Segmentation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19591","citing_title":"Structure-Semantic Decoupled Modulation of Global Geospatial Embeddings for High-Resolution Remote Sensing Mapping","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12319","citing_title":"RSGMamba: Reliability-Aware Self-Gated State Space Model for Multimodal Semantic Segmentation","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA","json":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA.json","graph_json":"https://pith.science/api/pith-number/HPYUXF6IAR46MJG4AF53AW5CIA/graph.json","events_json":"https://pith.science/api/pith-number/HPYUXF6IAR46MJG4AF53AW5CIA/events.json","paper":"https://pith.science/paper/HPYUXF6I"},"agent_actions":{"view_html":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA","download_json":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA.json","view_paper":"https://pith.science/paper/HPYUXF6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.01210&json=true","fetch_graph":"https://pith.science/api/pith-number/HPYUXF6IAR46MJG4AF53AW5CIA/graph.json","fetch_events":"https://pith.science/api/pith-number/HPYUXF6IAR46MJG4AF53AW5CIA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA/action/storage_attestation","attest_author":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA/action/author_attestation","sign_citation":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA/action/citation_signature","submit_replication":"https://pith.science/pith/HPYUXF6IAR46MJG4AF53AW5CIA/action/replication_record"}},"created_at":"2026-07-05T08:27:10.798237+00:00","updated_at":"2026-07-05T08:27:10.798237+00:00"}