{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/mmdu-a-multi-turn-multi-image-dialog","title":"MMDU: A Multi-Turn Multi-Image Dialog Understanding Benchmark and Instruction-Tuning Dataset for LVLMs","arxiv_id":"2406.11833","date":"2024-06-17","proceeding":null,"authors":["Ziyu Liu","Tao Chu","Yuhang Zang","Xilin Wei","Xiaoyi Dong","Pan Zhang","Zijian Liang","Yuanjun Xiong","Yu Qiao","Dahua Lin","Jiaqi Wang"],"abstract":"Generating natural and meaningful responses to communicate with multi-modal human inputs is a fundamental capability of Large Vision-Language Models(LVLMs). While current open-source LVLMs demonstrate promising performance in simplified scenarios such as single-turn single-image input, they fall short in real-world conversation scenarios such as following instructions in a long context history with multi-turn and multi-images. Existing LVLM benchmarks primarily focus on single-choice questions or short-form responses, which do not adequately assess the capabilities of LVLMs in real-world human-AI interaction applications. Therefore, we introduce MMDU, a comprehensive benchmark, and MMDU-45k, a large-scale instruction tuning dataset, designed to evaluate and improve LVLMs' abilities in multi-turn and multi-image conversations. We employ the clustering algorithm to ffnd the relevant images and textual descriptions from the open-source Wikipedia and construct the question-answer pairs by human annotators with the assistance of the GPT-4o model. MMDU has a maximum of 18k image+text tokens, 20 images, and 27 turns, which is at least 5x longer than previous benchmarks and poses challenges to current LVLMs. Our in-depth analysis of 15 representative LVLMs using MMDU reveals that open-source LVLMs lag behind closed-source counterparts due to limited conversational instruction tuning data. We demonstrate that ffne-tuning open-source LVLMs on MMDU-45k signiffcantly address this gap, generating longer and more accurate conversations, and improving scores on MMDU and existing benchmarks (MMStar: +1.1%, MathVista: +1.5%, ChartQA:+1.2%). Our contributions pave the way for bridging the gap between current LVLM models and real-world application demands. This project is available at https://github.com/Liuziyu77/MMDU.","url_abs":"https://arxiv.org/abs/2406.11833v2","url_pdf":"https://arxiv.org/pdf/2406.11833v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"mmdu-a-multi-turn-multi-image-dialog","repo_url":"https://github.com/liuziyu77/mmdu","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[{"slug":"mmdu","name":"MMDU","full_name":"Multi-Turn Multi-Image Dialog Understanding"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-question-answering-on-mm-vet","task":"Visual Question Answering","dataset":"MM-Vet","model":"InternLM-XC2 + MMDU-45k","rank_in_archive_order":122,"of":231,"metrics":{"GPT-4 score":"38.8"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.11833","atlas_url":"https://app.syntology.ai/?focus=2406.11833","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.11833"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/liuziyu77/mmdu","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":1,"ran_draft_wrong":1,"ran_fixture":2,"ran_honours":2,"ran_violates":1,"unverified":6},"by_repo_kind":{"official":{"samples":9,"ran":5,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":4,"samples":[{"code_sha256_prefix":"57e95166ade5f25c","entry":"build_transform","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/InternVL_chat_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/InternVL_chat_gen_ans.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"57e95166ade5f25c"}},{"code_sha256_prefix":"d610b5eabe0c9db1","entry":"dynamic_preprocess","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/InternVL_chat_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/InternVL_chat_gen_ans.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d610b5eabe0c9db1"}},{"code_sha256_prefix":"bd77f5f8067f18e9","entry":"find_closest_aspect_ratio","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/InternVL_chat_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/InternVL_chat_gen_ans.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bd77f5f8067f18e9"}},{"code_sha256_prefix":"9b3c1cb391672ccb","entry":"load_image","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/mmdu_LLaVa_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/mmdu_LLaVa_gen_ans.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9b3c1cb391672ccb"}},{"code_sha256_prefix":"d5044bdde33c54c2","entry":"load_images","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/mmdu_LLaVa_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/mmdu_LLaVa_gen_ans.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d5044bdde33c54c2"}},{"code_sha256_prefix":"763f6fe20bd52fbb","entry":"parse_score","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"763f6fe20bd52fbb"}},{"code_sha256_prefix":"10893c4608c08075","entry":"split_to_even_chunks","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"10893c4608c08075"}},{"code_sha256_prefix":"bb35e3ac741bb2c9","entry":"get_mm_adapter_state_maybe_zero_3","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"bb35e3ac741bb2c9"}},{"code_sha256_prefix":"735025744c1ab0cf","entry":"maybe_zero_3","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"735025744c1ab0cf"}},{"code_sha256_prefix":"6d3ac51d93b6fa7b","entry":"mllm_openai","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/Claude3opus_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/Claude3opus_gen_ans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6d3ac51d93b6fa7b"}},{"code_sha256_prefix":"9e43f38ac5b5a4f9","entry":"mllm_openai","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/GPT4o_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/GPT4o_gen_ans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9e43f38ac5b5a4f9"}},{"code_sha256_prefix":"def81a4cd2af1cad","entry":"mllm_openai","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/GPT4turbo_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/GPT4turbo_gen_ans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"def81a4cd2af1cad"}},{"code_sha256_prefix":"25182bd581875ec4","entry":"mllm_openai","repo":"liuziyu77/mmdu","repo_kind":"official","path":"model_generation/Qwen_VL_Max_gen_ans.py","file_url":"https://github.com/liuziyu77/mmdu/blob/HEAD/model_generation/Qwen_VL_Max_gen_ans.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"25182bd581875ec4"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}