{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:VK76W57UJ7BQTLIURIOM5RL5C6","short_pith_number":"pith:VK76W57U","schema_version":"1.0","canonical_sha256":"aabfeb77f44fc309ad148a1ccec57d17bd6de359a1ac04a13fcd56082cdc85db","source":{"kind":"arxiv","id":"2201.12086","version":2},"attestation_state":"computed","paper":{"title":"BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Caiming Xiong, Dongxu Li, Junnan Li, Steven Hoi","submitted_at":"2022-01-28T12:49:48Z","abstract_excerpt":"Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based tasks or generation-based tasks. Furthermore, performance improvement has been largely achieved by scaling up the dataset with noisy image-text pairs collected from the web, which is a suboptimal source of supervision. In this paper, we propose BLIP, a new VLP framework which transfers flexibly to both vision-language understanding and generation tasks. BLIP effectively utilizes the noisy web data by bootstrapping the"},"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":"2201.12086","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2022-01-28T12:49:48Z","cross_cats_sorted":[],"title_canon_sha256":"55c2083b78cdd879b4126ccc806bc030dea530747cbd767f4552963e178bf054","abstract_canon_sha256":"b1e393b0ed6306570df95f871cce2cd51723d8122d0156f27ea73a55ad4903c7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:56:54.279243Z","signature_b64":"rjGav3wGUGZSeZV2IW7C/CF+KnS1EtOEGds/Rqm5JJp2Ib8IINoIJeibRzbUywJanLlq69Rp6C00xFncOqU/BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aabfeb77f44fc309ad148a1ccec57d17bd6de359a1ac04a13fcd56082cdc85db","last_reissued_at":"2026-07-05T03:56:54.278823Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:56:54.278823Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Caiming Xiong, Dongxu Li, Junnan Li, Steven Hoi","submitted_at":"2022-01-28T12:49:48Z","abstract_excerpt":"Vision-Language Pre-training (VLP) has advanced the performance for many vision-language tasks. However, most existing pre-trained models only excel in either understanding-based tasks or generation-based tasks. Furthermore, performance improvement has been largely achieved by scaling up the dataset with noisy image-text pairs collected from the web, which is a suboptimal source of supervision. In this paper, we propose BLIP, a new VLP framework which transfers flexibly to both vision-language understanding and generation tasks. BLIP effectively utilizes the noisy web data by bootstrapping the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.12086","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/2201.12086/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":"2201.12086","created_at":"2026-07-05T03:56:54.278878+00:00"},{"alias_kind":"arxiv_version","alias_value":"2201.12086v2","created_at":"2026-07-05T03:56:54.278878+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.12086","created_at":"2026-07-05T03:56:54.278878+00:00"},{"alias_kind":"pith_short_12","alias_value":"VK76W57UJ7BQ","created_at":"2026-07-05T03:56:54.278878+00:00"},{"alias_kind":"pith_short_16","alias_value":"VK76W57UJ7BQTLIU","created_at":"2026-07-05T03:56:54.278878+00:00"},{"alias_kind":"pith_short_8","alias_value":"VK76W57U","created_at":"2026-07-05T03:56:54.278878+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.18347","citing_title":"Multilingual Training and Evaluation Resources for Vision-Language Models","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.24938","citing_title":"Your Embedding Model is SMARTer Than You Think","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30168","citing_title":"OmniCD: A Foundational Framework for Remote Sensing Image Change Detection Guided by Multimodal Semantics","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17669","citing_title":"Multimodal Cultural Heritage Knowledge Graph Extension with Language and Vision Models","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2511.00181","citing_title":"From Evidence to Verdict: An Agent-Based Forensic Framework for AI-Generated Image Detection","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2205.01917","citing_title":"CoCa: Contrastive Captioners are Image-Text Foundation Models","ref_index":37,"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":45,"is_internal_anchor":false},{"citing_arxiv_id":"2204.14198","citing_title":"Flamingo: a Visual Language Model for Few-Shot Learning","ref_index":59,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05900","citing_title":"AICA-Bench: Holistically Examining the Capabilities of VLMs in Affective Image Content Analysis","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05210","citing_title":"Integration of Object Detection and Small VLMs for Construction Safety Hazard Identification","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18347","citing_title":"Multilingual Training and Evaluation Resources for Vision-Language Models","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18167","citing_title":"Embedding Arithmetic: A Lightweight, Tuning-Free Framework for Post-hoc Bias Mitigation in Text-to-Image Models","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6","json":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6.json","graph_json":"https://pith.science/api/pith-number/VK76W57UJ7BQTLIURIOM5RL5C6/graph.json","events_json":"https://pith.science/api/pith-number/VK76W57UJ7BQTLIURIOM5RL5C6/events.json","paper":"https://pith.science/paper/VK76W57U"},"agent_actions":{"view_html":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6","download_json":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6.json","view_paper":"https://pith.science/paper/VK76W57U","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2201.12086&json=true","fetch_graph":"https://pith.science/api/pith-number/VK76W57UJ7BQTLIURIOM5RL5C6/graph.json","fetch_events":"https://pith.science/api/pith-number/VK76W57UJ7BQTLIURIOM5RL5C6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6/action/storage_attestation","attest_author":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6/action/author_attestation","sign_citation":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6/action/citation_signature","submit_replication":"https://pith.science/pith/VK76W57UJ7BQTLIURIOM5RL5C6/action/replication_record"}},"created_at":"2026-07-05T03:56:54.278878+00:00","updated_at":"2026-07-05T03:56:54.278878+00:00"}