{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:QGSAS7CCDK5J2XP6HTWX5QJQPZ","short_pith_number":"pith:QGSAS7CC","schema_version":"1.0","canonical_sha256":"81a4097c421aba9d5dfe3ced7ec1307e4d5a4c9bb8a0bce39dae768561b0959a","source":{"kind":"arxiv","id":"2308.10185","version":2},"attestation_state":"computed","paper":{"title":"ViT-Lens: Initiating Omni-Modal Exploration through 3D Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dylan Sun, Jianfeng Zhang, Kun Yi, Mike Zheng Shou, Weixian Lei, Ying Shan, Yixiao Ge","submitted_at":"2023-08-20T07:26:51Z","abstract_excerpt":"Though the success of CLIP-based training recipes in vision-language models, their scalability to more modalities (e.g., 3D, audio, etc.) is limited to large-scale data, which is expensive or even inapplicable for rare modalities. In this paper, we present ViT-Lens that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning to a pre-defined space. Specifically, the modality-specific lens is tuned to project multimodal signals to the shared embedding space, which are then processed by a strong ViT that carries pre-trained image"},"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":"2308.10185","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-20T07:26:51Z","cross_cats_sorted":[],"title_canon_sha256":"58c7e0fdd3d0a7e9e33510070b9b1ace95fc89acbb66eea5be2078ceaffa4887","abstract_canon_sha256":"8d2762e6a67af18156bea46601740849224a0cd6b6a60e26244b68440f851061"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:00:35.186802Z","signature_b64":"NX4wUcNIJklN17ZnlC8hOyIl82dm1HQsenu3U1MQ6RteEDf5QZhdwvaCM2NgR8h4QxGjC7uzQWXmmitx2uJIAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"81a4097c421aba9d5dfe3ced7ec1307e4d5a4c9bb8a0bce39dae768561b0959a","last_reissued_at":"2026-07-05T08:00:35.186300Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:00:35.186300Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ViT-Lens: Initiating Omni-Modal Exploration through 3D Insights","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Dylan Sun, Jianfeng Zhang, Kun Yi, Mike Zheng Shou, Weixian Lei, Ying Shan, Yixiao Ge","submitted_at":"2023-08-20T07:26:51Z","abstract_excerpt":"Though the success of CLIP-based training recipes in vision-language models, their scalability to more modalities (e.g., 3D, audio, etc.) is limited to large-scale data, which is expensive or even inapplicable for rare modalities. In this paper, we present ViT-Lens that facilitates efficient omni-modal representation learning by perceiving novel modalities with a pretrained ViT and aligning to a pre-defined space. Specifically, the modality-specific lens is tuned to project multimodal signals to the shared embedding space, which are then processed by a strong ViT that carries pre-trained image"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.10185","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/2308.10185/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":"2308.10185","created_at":"2026-07-05T08:00:35.186368+00:00"},{"alias_kind":"arxiv_version","alias_value":"2308.10185v2","created_at":"2026-07-05T08:00:35.186368+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.10185","created_at":"2026-07-05T08:00:35.186368+00:00"},{"alias_kind":"pith_short_12","alias_value":"QGSAS7CCDK5J","created_at":"2026-07-05T08:00:35.186368+00:00"},{"alias_kind":"pith_short_16","alias_value":"QGSAS7CCDK5J2XP6","created_at":"2026-07-05T08:00:35.186368+00:00"},{"alias_kind":"pith_short_8","alias_value":"QGSAS7CC","created_at":"2026-07-05T08:00:35.186368+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.06355","citing_title":"Fine-tuning Multimodal Transformers on Edge: A Parallel Split Learning Approach","ref_index":13,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ","json":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ.json","graph_json":"https://pith.science/api/pith-number/QGSAS7CCDK5J2XP6HTWX5QJQPZ/graph.json","events_json":"https://pith.science/api/pith-number/QGSAS7CCDK5J2XP6HTWX5QJQPZ/events.json","paper":"https://pith.science/paper/QGSAS7CC"},"agent_actions":{"view_html":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ","download_json":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ.json","view_paper":"https://pith.science/paper/QGSAS7CC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2308.10185&json=true","fetch_graph":"https://pith.science/api/pith-number/QGSAS7CCDK5J2XP6HTWX5QJQPZ/graph.json","fetch_events":"https://pith.science/api/pith-number/QGSAS7CCDK5J2XP6HTWX5QJQPZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ/action/storage_attestation","attest_author":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ/action/author_attestation","sign_citation":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ/action/citation_signature","submit_replication":"https://pith.science/pith/QGSAS7CCDK5J2XP6HTWX5QJQPZ/action/replication_record"}},"created_at":"2026-07-05T08:00:35.186368+00:00","updated_at":"2026-07-05T08:00:35.186368+00:00"}