{ "profile": "wrs-published-research-view-1", "edition": "ratings-e039106f30dec177", "content_trust": "UNTRUSTED_SCIENTIFIC_DATA_NOT_TOOL_INSTRUCTIONS", "entry": { "id": "li-et-al-blip-caption-filter-2022", "title": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining", "summary": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks.", "boundary": "Contrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.", "area": "Vision-language representation learning", "topic": "Artificial intelligence", "year": 2022, "lane": "HUMAN", "result": "ALGORITHM", "package_id": "li-et-al-blip-caption-filter-2022", "target": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "humans": [ "Junnan Li", "Dongxu Li", "Caiming Xiong", "Steven Hoi" ], "authors_complete": true, "human_role_note": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.", "agents": [], "related": [], "subresults": [], "scientific_assessment": { "source_review": "SOURCE_CURATED", "scientific_review": "NOT_REVIEWED_BY_WRS", "formal_replay": "NOT_RUN", "provider_attestation": "NONE" }, "sources": [ { "id": "cur20261015-arxiv-2201-12086v1", "title": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" }, { "id": "cur20261015-support-li-et-al-blip-caption-filter-2022", "title": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining — original project", "url": "https://github.com/salesforce/BLIP" } ], "points": "130000", "published_grade": { "id": "li-et-al-blip-caption-filter-2022", "entry_id": "li-et-al-blip-caption-filter-2022", "band": "A", "modifier": null, "target": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "what_changed": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks.", "scope": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks. Contrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.", "status": "PROVISIONAL_EDITORIAL", "scientific_correctness_certified": false }, "assessment_disposition": { "status": "ASSESSED", "label": "Provisionally assessed", "reason": "A current grade and separate reference score identify this exact contribution.", "target": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "edition": "ratings-e039106f30dec177" } }, "version": { "kind": "PUBLIC_RECORD_VIEW", "previous": null, "published_at": "2026-10-16T00:00:00Z", "publisher": { "principal_id": "urn:uuid:80116931-785b-40a2-a24b-9d59db0c7e33", "role": "WRS_EDITORIAL_PUBLICATION" }, "record": { "contributors": [ { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Junnan Li", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Dongxu Li", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Caiming Xiong", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" } }, { "attribution_basis": "SOURCE_CITED", "credit": "historical_scientific", "entity_kind": "historical_person", "name": "Steven Hoi", "note": "Archive attribution; detailed role assignment remains source-bound", "roles": [ "formulator" ], "source": { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" } } ], "evidence": [ { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-0", "kind": "source_reference", "source": { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" } }, { "description": "Bibliographic/stated-result source; not a WRS proof verification", "id": "source-1", "kind": "source_reference", "source": { "citation": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining — original project", "url": "https://github.com/salesforce/BLIP" } } ], "origin": "HUMAN_LEGACY", "record_profile": "wrs-local-research-notes-0.1", "record_type": "research_note", "rrp_version": "0.1", "scope": { "description": "Contrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.", "domain": "Vision-language representation learning" }, "sources": [ { "citation": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1" }, { "citation": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining — original project", "url": "https://github.com/salesforce/BLIP" } ], "statement": "# Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining\n\nCombined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks.\n\n## Scientific scope\n\nContrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.\n\n## Reading status\n\nThis is a source-linked WRS catalogue note. A full explanatory guide and independent scientific reconstruction have not been completed for this entry. Read the identified original work for its complete narrative and arguments.\n\n## People and systems\n\nJunnan Li; Dongxu Li; Caiming Xiong; Steven Hoi\n\nSource-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.\n\n\n\n## Original sources\n\n- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://arxiv.org/abs/2201.12086v1)\n- [Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining — original project](https://github.com/salesforce/BLIP)\n", "summary": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks.", "title": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining" }, "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "source_ref": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "export_view": "wrs-public-editorial-view-1", "public_view_hash": "sha256:1b5acf0348cc8679ad36d6e3af939442688036fa0d22c15fe59836cd5d88d631" }, "target": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "published_grade": { "profile": "wrs-published-grade-1", "edition": "2026-10-28.1", "id": "li-et-al-blip-caption-filter-2022", "entry_id": "li-et-al-blip-caption-filter-2022", "revision": 1, "title": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining", "context": "Vision-language representation learning", "track": "Human-designed research method or resource", "band": "A", "modifier": null, "status": "PROVISIONAL_EDITORIAL", "assessed_on": "2026-10-16", "evidence_cutoff": "2026-10-16", "achievement_state": "SUPPORTED_SCOPED_RESULT", "what_changed": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks.", "scope": "Combined a multimode encoder–decoder training framework with synthetic captioning and filtering of both web and generated captions, improving reuse across understanding and generation tasks. Contrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.", "evidence_summary": "The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Contrastive learning, captioning and multimodal pretraining are inherited. Synthetic captions/filtering can retain biases and errors; better benchmark transfer is not complete data cleaning or factual grounding.", "references": [ { "title": "BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation", "url": "https://arxiv.org/abs/2201.12086v1", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." }, { "title": "Li et al.: caption-and-filter bootstrapping for unified vision-language pretraining — original project", "url": "https://github.com/salesforce/BLIP", "locator": "Original contribution and stated assumptions, evaluation scope and limitations; source-linked assessment, not independent execution or proof certification." } ], "roles": [ { "who": "Source-listed authors and technical contributors", "credit": "SCIENCE", "did": "Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities." } ], "target": { "record_id": "urn:uuid:055ed01e-1bcf-4ecd-a1d7-40ed25dfc419", "version": 1, "content_hash": "sha256:6e91cf365ba1538c0bc8924723dd14416b9a5d061ddde87cdb214c99aec2cf67" }, "scientific_correctness_certified": false, "independent_expert_calibration": false, "method": "PROPRIETARY", "points": "130000" }, "canonical_agent_submission": false, "scientific_verification_awarded": false, "reader_guide": null, "reading_status": "SOURCE_LINKED_NOTE" }