{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:65ZNJHXBKAEA2YESAVR3UU5ON2","short_pith_number":"pith:65ZNJHXB","schema_version":"1.0","canonical_sha256":"f772d49ee150080d60920563ba53ae6e870ee96e838a29f502687c8b28e1c9d4","source":{"kind":"arxiv","id":"2504.09518","version":1},"attestation_state":"computed","paper":{"title":"3D CoCa: Contrastive Learners are 3D Captioners","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Tang, Ting Huang, Yemin Wang, Zeyu Zhang","submitted_at":"2025-04-13T11:10:47Z","abstract_excerpt":"3D captioning, which aims to describe the content of 3D scenes in natural language, remains highly challenging due to the inherent sparsity of point clouds and weak cross-modal alignment in existing methods. To address these challenges, we propose 3D CoCa, a novel unified framework that seamlessly combines contrastive vision-language learning with 3D caption generation in a single architecture. Our approach leverages a frozen CLIP vision-language backbone to provide rich semantic priors, a spatially-aware 3D scene encoder to capture geometric context, and a multi-modal decoder to generate desc"},"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":"2504.09518","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-13T11:10:47Z","cross_cats_sorted":[],"title_canon_sha256":"927dece04fb9195006673ad7275417a280aaff3cb53ba7f64fecd572ae1597b9","abstract_canon_sha256":"0740863111df0172d880b27ab4c06e0c2b86b4180d1c23f86f1d1dc68b94f507"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:48:38.942198Z","signature_b64":"VimiUyS1dMu7VEUWAj9QsjQqPenqAS7kCe0m+D1jxc2GmYfLsQvqzhgkQ8NxGj2bOOQqB3i6Zglwvctosvn8Cg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f772d49ee150080d60920563ba53ae6e870ee96e838a29f502687c8b28e1c9d4","last_reissued_at":"2026-07-05T10:48:38.941735Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:48:38.941735Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"3D CoCa: Contrastive Learners are 3D Captioners","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hao Tang, Ting Huang, Yemin Wang, Zeyu Zhang","submitted_at":"2025-04-13T11:10:47Z","abstract_excerpt":"3D captioning, which aims to describe the content of 3D scenes in natural language, remains highly challenging due to the inherent sparsity of point clouds and weak cross-modal alignment in existing methods. To address these challenges, we propose 3D CoCa, a novel unified framework that seamlessly combines contrastive vision-language learning with 3D caption generation in a single architecture. Our approach leverages a frozen CLIP vision-language backbone to provide rich semantic priors, a spatially-aware 3D scene encoder to capture geometric context, and a multi-modal decoder to generate desc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.09518","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/2504.09518/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":"2504.09518","created_at":"2026-07-05T10:48:38.941793+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.09518v1","created_at":"2026-07-05T10:48:38.941793+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.09518","created_at":"2026-07-05T10:48:38.941793+00:00"},{"alias_kind":"pith_short_12","alias_value":"65ZNJHXBKAEA","created_at":"2026-07-05T10:48:38.941793+00:00"},{"alias_kind":"pith_short_16","alias_value":"65ZNJHXBKAEA2YES","created_at":"2026-07-05T10:48:38.941793+00:00"},{"alias_kind":"pith_short_8","alias_value":"65ZNJHXB","created_at":"2026-07-05T10:48:38.941793+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.17472","citing_title":"UniMesh: Unifying 3D Mesh Understanding and Generation","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2","json":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2.json","graph_json":"https://pith.science/api/pith-number/65ZNJHXBKAEA2YESAVR3UU5ON2/graph.json","events_json":"https://pith.science/api/pith-number/65ZNJHXBKAEA2YESAVR3UU5ON2/events.json","paper":"https://pith.science/paper/65ZNJHXB"},"agent_actions":{"view_html":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2","download_json":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2.json","view_paper":"https://pith.science/paper/65ZNJHXB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.09518&json=true","fetch_graph":"https://pith.science/api/pith-number/65ZNJHXBKAEA2YESAVR3UU5ON2/graph.json","fetch_events":"https://pith.science/api/pith-number/65ZNJHXBKAEA2YESAVR3UU5ON2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2/action/storage_attestation","attest_author":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2/action/author_attestation","sign_citation":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2/action/citation_signature","submit_replication":"https://pith.science/pith/65ZNJHXBKAEA2YESAVR3UU5ON2/action/replication_record"}},"created_at":"2026-07-05T10:48:38.941793+00:00","updated_at":"2026-07-05T10:48:38.941793+00:00"}