{"paper":{"title":"A Survey on Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multimodal large language models use LLMs as a central brain to handle images and other inputs with new emergent reasoning skills.","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Chaoyou Fu, Enhong Chen, Ke Li, Shukang Yin, Sirui Zhao, Tong Xu, Xing Sun","submitted_at":"2023-06-23T15:21:52Z","abstract_excerpt":"Recently, Multimodal Large Language Model (MLLM) represented by GPT-4V has been a new rising research hotspot, which uses powerful Large Language Models (LLMs) as a brain to perform multimodal tasks. The surprising emergent capabilities of MLLM, such as writing stories based on images and OCR-free math reasoning, are rare in traditional multimodal methods, suggesting a potential path to artificial general intelligence. To this end, both academia and industry have endeavored to develop MLLMs that can compete with or even better than GPT-4V, pushing the limit of research at a surprising speed. I"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The surprising emergent capabilities of MLLM, such as writing stories based on images and OCR-free math reasoning, are rare in traditional multimodal methods, suggesting a potential path to artificial general intelligence.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The survey assumes that the cited literature and the associated GitHub repository together provide a sufficiently complete and up-to-date picture of the rapidly evolving MLLM field.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"This survey organizes the architectures, training strategies, data, evaluation methods, extensions, and challenges of Multimodal Large Language Models.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Multimodal large language models use LLMs as a central brain to handle images and other inputs with new emergent reasoning skills.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a3fb4de2954e150bf7db9f22a38bc6a1ff5afdb7e16ca4a708d2b0819c75b40c"},"source":{"id":"2306.13549","kind":"arxiv","version":4},"verdict":{"id":"e49d6b2a-fdae-4044-a620-a884a3578c51","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T02:48:44.972565Z","strongest_claim":"The surprising emergent capabilities of MLLM, such as writing stories based on images and OCR-free math reasoning, are rare in traditional multimodal methods, suggesting a potential path to artificial general intelligence.","one_line_summary":"This survey organizes the architectures, training strategies, data, evaluation methods, extensions, and challenges of Multimodal Large Language Models.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The survey assumes that the cited literature and the associated GitHub repository together provide a sufficiently complete and up-to-date picture of the rapidly evolving MLLM field.","pith_extraction_headline":"Multimodal large language models use LLMs as a central brain to handle images and other inputs with new emergent reasoning skills."},"references":{"count":209,"sample":[{"doi":"","year":2023,"title":"A Survey of Large Language Models","work_id":"de1b42b5-4a0a-4b1f-8c78-1f7fe21be6c9","ref_index":1,"cited_arxiv_id":"2303.18223","is_internal_anchor":true},{"doi":"","year":2023,"title":"Chatgpt: A language model for conversational ai,","work_id":"5c8bde1b-9044-47c0-9f75-2681238b7fc4","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"GPT-4 Technical Report","work_id":"b928e041-6991-4c08-8c81-0359e4097c7b","ref_index":3,"cited_arxiv_id":"2303.08774","is_internal_anchor":true},{"doi":"","year":null,"title":"Vicuna: An open-source chatbot impressing gpt-4 with 90% chatgpt quality,","work_id":"1c1e30f3-a043-4016-b0e5-66862973f8b2","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":null,"title":"Available: https://vicuna.lmsys.org 1, 3, 4","work_id":"d25bb515-3b10-4c5d-bf60-9cc603177a24","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":209,"snapshot_sha256":"9d83048368ed1998ccb929e20a79d5b839c83ee3e5a9fd1d135b600e0449c33f","internal_anchors":72},"formal_canon":{"evidence_count":2,"snapshot_sha256":"40d027c84c5429daaf05d2ff534b9893bd2db24ea618e012f51051da244dc3d3"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}