{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:7YUNHCRRQ5G5INXQO3NIBNT4ZK","short_pith_number":"pith:7YUNHCRR","schema_version":"1.0","canonical_sha256":"fe28d38a31874dd436f076da80b67cca9f4aa384dfb8797e0c758ae46c4b749c","source":{"kind":"arxiv","id":"2107.06383","version":1},"attestation_state":"computed","paper":{"title":"How Much Can CLIP Benefit Vision-and-Language Tasks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anna Rohrbach, Hao Tan, Kai-Wei Chang, Kurt Keutzer, Liunian Harold Li, Mohit Bansal, Sheng Shen, Zhewei Yao","submitted_at":"2021-07-13T20:48:12Z","abstract_excerpt":"Most existing Vision-and-Language (V&L) models rely on pre-trained visual encoders, using a relatively small set of manually-annotated data (as compared to web-crawled data), to perceive the visual world. However, it has been observed that large-scale pretraining usually can result in better generalization performance, e.g., CLIP (Contrastive Language-Image Pre-training), trained on a massive amount of image-caption pairs, has shown a strong zero-shot capability on various vision tasks. To further study the advantage brought by CLIP, we propose to use CLIP as the visual encoder in various V&L "},"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":"2107.06383","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-07-13T20:48:12Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"4954c41cc5444b065841840cf6ef5811f6fc359fa18fb6eec942d33024354257","abstract_canon_sha256":"6ad9c501178ec31d787ae6732d2927ae6b58d67793fa15b9334faa73d6074fe6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:57:57.489760Z","signature_b64":"Vgs2BDgDVmZrW9JGapFo+6BmM/GClbGeCmG7gfH1KZ65sDgGTif5HI5vGjLeeZVTMbaOOz0LK1X88rQpBqRVDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fe28d38a31874dd436f076da80b67cca9f4aa384dfb8797e0c758ae46c4b749c","last_reissued_at":"2026-07-05T02:57:57.489264Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:57:57.489264Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Much Can CLIP Benefit Vision-and-Language Tasks?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Anna Rohrbach, Hao Tan, Kai-Wei Chang, Kurt Keutzer, Liunian Harold Li, Mohit Bansal, Sheng Shen, Zhewei Yao","submitted_at":"2021-07-13T20:48:12Z","abstract_excerpt":"Most existing Vision-and-Language (V&L) models rely on pre-trained visual encoders, using a relatively small set of manually-annotated data (as compared to web-crawled data), to perceive the visual world. However, it has been observed that large-scale pretraining usually can result in better generalization performance, e.g., CLIP (Contrastive Language-Image Pre-training), trained on a massive amount of image-caption pairs, has shown a strong zero-shot capability on various vision tasks. To further study the advantage brought by CLIP, we propose to use CLIP as the visual encoder in various V&L "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.06383","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/2107.06383/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":"2107.06383","created_at":"2026-07-05T02:57:57.489330+00:00"},{"alias_kind":"arxiv_version","alias_value":"2107.06383v1","created_at":"2026-07-05T02:57:57.489330+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.06383","created_at":"2026-07-05T02:57:57.489330+00:00"},{"alias_kind":"pith_short_12","alias_value":"7YUNHCRRQ5G5","created_at":"2026-07-05T02:57:57.489330+00:00"},{"alias_kind":"pith_short_16","alias_value":"7YUNHCRRQ5G5INXQ","created_at":"2026-07-05T02:57:57.489330+00:00"},{"alias_kind":"pith_short_8","alias_value":"7YUNHCRR","created_at":"2026-07-05T02:57:57.489330+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2212.03191","citing_title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2205.14100","citing_title":"GIT: A Generative Image-to-text Transformer for Vision and Language","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2601.09746","citing_title":"Multi-Agent Cooperative Learning for Robust Vision-Language Alignment under OOD Concepts","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2401.01614","citing_title":"GPT-4V(ision) is a Generalist Web Agent, if Grounded","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2205.01917","citing_title":"CoCa: Contrastive Captioners are Image-Text Foundation Models","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2307.06942","citing_title":"InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2305.06355","citing_title":"VideoChat: Chat-Centric Video Understanding","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2210.08402","citing_title":"LAION-5B: An open large-scale dataset for training next generation image-text models","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09088","citing_title":"Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation","ref_index":80,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04058","citing_title":"MP-ISMoE: Mixed-Precision Interactive Side Mixture-of-Experts for Efficient Transfer Learning","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05809","citing_title":"Adjustable Text-Guided Backdoor Attacks with Natural-Word Triggers on Multimodal Pretrained Models","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2204.06125","citing_title":"Hierarchical Text-Conditional Image Generation with CLIP Latents","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17472","citing_title":"UniMesh: Unifying 3D Mesh Understanding and Generation","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK","json":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK.json","graph_json":"https://pith.science/api/pith-number/7YUNHCRRQ5G5INXQO3NIBNT4ZK/graph.json","events_json":"https://pith.science/api/pith-number/7YUNHCRRQ5G5INXQO3NIBNT4ZK/events.json","paper":"https://pith.science/paper/7YUNHCRR"},"agent_actions":{"view_html":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK","download_json":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK.json","view_paper":"https://pith.science/paper/7YUNHCRR","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2107.06383&json=true","fetch_graph":"https://pith.science/api/pith-number/7YUNHCRRQ5G5INXQO3NIBNT4ZK/graph.json","fetch_events":"https://pith.science/api/pith-number/7YUNHCRRQ5G5INXQO3NIBNT4ZK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK/action/storage_attestation","attest_author":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK/action/author_attestation","sign_citation":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK/action/citation_signature","submit_replication":"https://pith.science/pith/7YUNHCRRQ5G5INXQO3NIBNT4ZK/action/replication_record"}},"created_at":"2026-07-05T02:57:57.489330+00:00","updated_at":"2026-07-05T02:57:57.489330+00:00"}