{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:ZP43MHL435TGN2A5YLTAMW7QD3","short_pith_number":"pith:ZP43MHL4","schema_version":"1.0","canonical_sha256":"cbf9b61d7cdf6666e81dc2e6065bf01ee73f3e663e84c87ce44fcdc3a4772dfc","source":{"kind":"arxiv","id":"2305.14018","version":2},"attestation_state":"computed","paper":{"title":"Sparse4D v2: Recurrent Temporal Fusion with Sparse Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lichao Huang, Tianwei Lin, Xuewu Lin, Zhizhong Su, Zixiang Pei","submitted_at":"2023-05-23T12:53:58Z","abstract_excerpt":"Sparse algorithms offer great flexibility for multi-view temporal perception tasks. In this paper, we present an enhanced version of Sparse4D, in which we improve the temporal fusion module by implementing a recursive form of multi-frame feature sampling. By effectively decoupling image features and structured anchor features, Sparse4D enables a highly efficient transformation of temporal features, thereby facilitating temporal fusion solely through the frame-by-frame transmission of sparse features. The recurrent temporal fusion approach provides two main benefits. Firstly, it reduces the com"},"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":"2305.14018","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-05-23T12:53:58Z","cross_cats_sorted":[],"title_canon_sha256":"d2fa2e175fd05e032740045329f697fbcbd26d5f6ccc435ad9e19e9e8875535b","abstract_canon_sha256":"e19146d6cd45b92249b18feb7cf8347b385bf7b6a824dd065cd67e3a488c1a7c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:13:21.773766Z","signature_b64":"TW7STdyXDiEUNJvX0wBkADHA2OvsaZEwITHu4NZFfVRyKSpup/teQRrI/Ih7tKLbVInmwzrzi9p9eMByr/dhDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbf9b61d7cdf6666e81dc2e6065bf01ee73f3e663e84c87ce44fcdc3a4772dfc","last_reissued_at":"2026-07-05T06:13:21.773332Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:13:21.773332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparse4D v2: Recurrent Temporal Fusion with Sparse Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lichao Huang, Tianwei Lin, Xuewu Lin, Zhizhong Su, Zixiang Pei","submitted_at":"2023-05-23T12:53:58Z","abstract_excerpt":"Sparse algorithms offer great flexibility for multi-view temporal perception tasks. In this paper, we present an enhanced version of Sparse4D, in which we improve the temporal fusion module by implementing a recursive form of multi-frame feature sampling. By effectively decoupling image features and structured anchor features, Sparse4D enables a highly efficient transformation of temporal features, thereby facilitating temporal fusion solely through the frame-by-frame transmission of sparse features. The recurrent temporal fusion approach provides two main benefits. Firstly, it reduces the com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.14018","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/2305.14018/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":"2305.14018","created_at":"2026-07-05T06:13:21.773388+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.14018v2","created_at":"2026-07-05T06:13:21.773388+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.14018","created_at":"2026-07-05T06:13:21.773388+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZP43MHL435TG","created_at":"2026-07-05T06:13:21.773388+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZP43MHL435TGN2A5","created_at":"2026-07-05T06:13:21.773388+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZP43MHL4","created_at":"2026-07-05T06:13:21.773388+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.14110","citing_title":"SToRe3D: Sparse Token Relevance in ViTs for Efficient Multi-View 3D Object Detection","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2604.00813","citing_title":"DVGT-2: Vision-Geometry-Action Model for Autonomous Driving at Scale","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01924","citing_title":"SimPB++: Simultaneously Detecting 2D and 3D Objects from Multiple Cameras","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05449","citing_title":"Not All Agents Matter: From Global Attention Dilution to Risk-Prioritized Game Planning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13586","citing_title":"Efficient Multi-View 3D Object Detection by Dynamic Token Selection and Fine-Tuning","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14563","citing_title":"Revisiting Token Compression for Accelerating ViT-based Sparse Multi-View 3D Object Detectors","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17024","citing_title":"CAM3DNet: Comprehensively mining the multi-scale features for 3D Object Detection with Multi-View Cameras","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18476","citing_title":"SemLT3D: Semantic-Guided Expert Distillation for Camera-only Long-Tailed 3D Object Detection","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3","json":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3.json","graph_json":"https://pith.science/api/pith-number/ZP43MHL435TGN2A5YLTAMW7QD3/graph.json","events_json":"https://pith.science/api/pith-number/ZP43MHL435TGN2A5YLTAMW7QD3/events.json","paper":"https://pith.science/paper/ZP43MHL4"},"agent_actions":{"view_html":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3","download_json":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3.json","view_paper":"https://pith.science/paper/ZP43MHL4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.14018&json=true","fetch_graph":"https://pith.science/api/pith-number/ZP43MHL435TGN2A5YLTAMW7QD3/graph.json","fetch_events":"https://pith.science/api/pith-number/ZP43MHL435TGN2A5YLTAMW7QD3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3/action/storage_attestation","attest_author":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3/action/author_attestation","sign_citation":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3/action/citation_signature","submit_replication":"https://pith.science/pith/ZP43MHL435TGN2A5YLTAMW7QD3/action/replication_record"}},"created_at":"2026-07-05T06:13:21.773388+00:00","updated_at":"2026-07-05T06:13:21.773388+00:00"}