{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZYTAPYSC2TCIMOFIMRMIAHR3HG","short_pith_number":"pith:ZYTAPYSC","schema_version":"1.0","canonical_sha256":"ce2607e242d4c48638a86458801e3b3993a1bdc054f381c7e9f11dda2498a8dd","source":{"kind":"arxiv","id":"2208.10442","version":2},"attestation_state":"computed","paper":{"title":"Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Furu Wei, Hangbo Bao, Johan Bjorck, Kriti Aggarwal, Li Dong, Owais Khan Mohammed, Qiang Liu, Saksham Singhal, Subhojit Som, Wenhui Wang, Zhiliang Peng","submitted_at":"2022-08-22T16:55:04Z","abstract_excerpt":"A big convergence of language, vision, and multimodal pretraining is emerging. In this work, we introduce a general-purpose multimodal foundation model BEiT-3, which achieves state-of-the-art transfer performance on both vision and vision-language tasks. Specifically, we advance the big convergence from three aspects: backbone architecture, pretraining task, and model scaling up. We introduce Multiway Transformers for general-purpose modeling, where the modular architecture enables both deep fusion and modality-specific encoding. Based on the shared backbone, we perform masked \"language\" model"},"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":"2208.10442","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2022-08-22T16:55:04Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f172fd4ec3481cf4290bf19bacdb4b4f6b9d723a5cf013793f321e357d471781","abstract_canon_sha256":"9d08d1aea34bc27006cbf08bf99983d5a35121f72abcb8ed87b598a57f90f6aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:53:30.025499Z","signature_b64":"Q1KOsSESMtnamqT5DVWVezLI+3AhYCpxb+dkQGR+w3wNNb3icZlLH7Ec8gPbq+ssFTekSueQXBTvrQXRhXQWBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce2607e242d4c48638a86458801e3b3993a1bdc054f381c7e9f11dda2498a8dd","last_reissued_at":"2026-07-05T04:53:30.025098Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:53:30.025098Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Image as a Foreign Language: BEiT Pretraining for All Vision and Vision-Language Tasks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Furu Wei, Hangbo Bao, Johan Bjorck, Kriti Aggarwal, Li Dong, Owais Khan Mohammed, Qiang Liu, Saksham Singhal, Subhojit Som, Wenhui Wang, Zhiliang Peng","submitted_at":"2022-08-22T16:55:04Z","abstract_excerpt":"A big convergence of language, vision, and multimodal pretraining is emerging. In this work, we introduce a general-purpose multimodal foundation model BEiT-3, which achieves state-of-the-art transfer performance on both vision and vision-language tasks. Specifically, we advance the big convergence from three aspects: backbone architecture, pretraining task, and model scaling up. We introduce Multiway Transformers for general-purpose modeling, where the modular architecture enables both deep fusion and modality-specific encoding. Based on the shared backbone, we perform masked \"language\" model"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.10442","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/2208.10442/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":"2208.10442","created_at":"2026-07-05T04:53:30.025153+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.10442v2","created_at":"2026-07-05T04:53:30.025153+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.10442","created_at":"2026-07-05T04:53:30.025153+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYTAPYSC2TCI","created_at":"2026-07-05T04:53:30.025153+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYTAPYSC2TCIMOFI","created_at":"2026-07-05T04:53:30.025153+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYTAPYSC","created_at":"2026-07-05T04:53:30.025153+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":16,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17030","citing_title":"Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation","ref_index":116,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09853","citing_title":"SynIB: Informational Bottleneck for Maximizing Synergy in Multimodal Learning","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2406.05410","citing_title":"ChatSR: Multimodal Large Language Models for Scientific Formula Discovery","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2402.03766","citing_title":"MobileVLM V2: Faster and Stronger Baseline for Vision Language Model","ref_index":67,"is_internal_anchor":false},{"citing_arxiv_id":"2509.22123","citing_title":"Multilingual Vision-Language Models, A Survey","ref_index":150,"is_internal_anchor":false},{"citing_arxiv_id":"2411.04996","citing_title":"Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2305.18565","citing_title":"PaLI-X: On Scaling up a Multilingual Vision and Language Model","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2212.03191","citing_title":"InternVideo: General Video Foundation Models via Generative and Discriminative Learning","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2209.06794","citing_title":"PaLI: A Jointly-Scaled Multilingual Language-Image Model","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2401.15947","citing_title":"MoE-LLaVA: Mixture of Experts for Large Vision-Language Models","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2302.14045","citing_title":"Language Is Not All You Need: Aligning Perception with Language Models","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2303.16199","citing_title":"LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention","ref_index":142,"is_internal_anchor":false},{"citing_arxiv_id":"2309.16588","citing_title":"Vision Transformers Need Registers","ref_index":280,"is_internal_anchor":false},{"citing_arxiv_id":"2407.07726","citing_title":"PaliGemma: A versatile 3B VLM for transfer","ref_index":141,"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":62,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18562","citing_title":"AnchorSeg: Language Grounded Query Banks for Reasoning Segmentation","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG","json":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG.json","graph_json":"https://pith.science/api/pith-number/ZYTAPYSC2TCIMOFIMRMIAHR3HG/graph.json","events_json":"https://pith.science/api/pith-number/ZYTAPYSC2TCIMOFIMRMIAHR3HG/events.json","paper":"https://pith.science/paper/ZYTAPYSC"},"agent_actions":{"view_html":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG","download_json":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG.json","view_paper":"https://pith.science/paper/ZYTAPYSC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.10442&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYTAPYSC2TCIMOFIMRMIAHR3HG/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYTAPYSC2TCIMOFIMRMIAHR3HG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG/action/storage_attestation","attest_author":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG/action/author_attestation","sign_citation":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG/action/citation_signature","submit_replication":"https://pith.science/pith/ZYTAPYSC2TCIMOFIMRMIAHR3HG/action/replication_record"}},"created_at":"2026-07-05T04:53:30.025153+00:00","updated_at":"2026-07-05T04:53:30.025153+00:00"}