{"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/mvbench-a-comprehensive-multi-modal-video","title":"MVBench: A Comprehensive Multi-modal Video Understanding Benchmark","arxiv_id":"2311.17005","date":"2023-11-28","proceeding":"CVPR 2024 1","authors":["Kunchang Li","Yali Wang","Yinan He","Yizhuo Li","Yi Wang","Yi Liu","Zun Wang","Jilan Xu","Guo Chen","Ping Luo","LiMin Wang","Yu Qiao"],"abstract":"With the rapid development of Multi-modal Large Language Models (MLLMs), a number of diagnostic benchmarks have recently emerged to evaluate the comprehension capabilities of these models. However, most benchmarks predominantly assess spatial understanding in the static image tasks, while overlooking temporal understanding in the dynamic video tasks. To alleviate this issue, we introduce a comprehensive Multi-modal Video understanding Benchmark, namely MVBench, which covers 20 challenging video tasks that cannot be effectively solved with a single frame. Specifically, we first introduce a novel static-to-dynamic method to define these temporal-related tasks. By transforming various static tasks into dynamic ones, we enable the systematic generation of video tasks that require a broad spectrum of temporal skills, ranging from perception to cognition. Then, guided by the task definition, we automatically convert public video annotations into multiple-choice QA to evaluate each task. On one hand, such a distinct paradigm allows us to build MVBench efficiently, without much manual intervention. On the other hand, it guarantees evaluation fairness with ground-truth video annotations, avoiding the biased scoring of LLMs. Moreover, we further develop a robust video MLLM baseline, i.e., VideoChat2, by progressive multi-modal training with diverse instruction-tuning data. The extensive results on our MVBench reveal that, the existing MLLMs are far from satisfactory in temporal understanding, while our VideoChat2 largely surpasses these leading models by over 15% on MVBench. All models and data are available at https://github.com/OpenGVLab/Ask-Anything.","url_abs":"https://arxiv.org/abs/2311.17005v4","url_pdf":"https://arxiv.org/pdf/2311.17005v4.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":"mvbench-a-comprehensive-multi-modal-video","repo_url":"https://github.com/opengvlab/ask-anything","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"mvbench-a-comprehensive-multi-modal-video","repo_url":"https://github.com/bytedance/tarsier","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"mvbench-a-comprehensive-multi-modal-video","repo_url":"https://github.com/magic-research/PLLaVA","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"3d-question-answering-3d-qa","task_name":"3D Question Answering (3D-QA)"},{"task_slug":"diagnostic","task_name":"Diagnostic"},{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":null,"task_name":"MVBench"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"vcgbench-diverse","task_name":"VCGBench-Diverse"},{"task_slug":"video-question-answering","task_name":"Video Question Answering"},{"task_slug":"video-understanding","task_name":"Video Understanding"},{"task_slug":"video-based-generative-performance","task_name":"Video-based Generative Performance Benchmarking"},{"task_slug":"video-based-generative-performance-5","task_name":"Video-based Generative Performance Benchmarking (Consistency)"},{"task_slug":"video-based-generative-performance-3","task_name":"Video-based Generative Performance Benchmarking (Contextual Understanding)"},{"task_slug":"video-based-generative-performance-1","task_name":"Video-based Generative Performance Benchmarking (Correctness of Information)"},{"task_slug":"video-based-generative-performance-2","task_name":"Video-based Generative Performance Benchmarking (Detail Orientation))"},{"task_slug":"video-based-generative-performance-4","task_name":"Video-based Generative Performance Benchmarking (Temporal Understanding)"},{"task_slug":"zero-shot-learning","task_name":"Zero-Shot Learning"},{"task_slug":"zeroshot-video-question-answer","task_name":"Zero-Shot Video Question Answer"}],"methods":[],"datasets_introduced":[{"slug":"mvbench","name":"MVBench","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-question-answering-3d-qa-on-sqa3d","task":"3D Question Answering (3D-QA)","dataset":"SQA3D","model":"VideoChat2","rank_in_archive_order":12,"of":13,"metrics":{"Exact Match":"37.3"},"uses_additional_data":false},{"leaderboard":"/sota/3d-question-answering-3d-qa-on-scanqa-test-w","task":"3D Question Answering (3D-QA)","dataset":"ScanQA Test w/ objects","model":"VideoChat2","rank_in_archive_order":13,"of":18,"metrics":{"BLEU-4":"9.6","CIDEr":"49.2","Exact Match":"19.2","METEOR":"9.5","ROUGE":"28.2"},"uses_additional_data":false},{"leaderboard":"/sota/vcgbench-diverse-on-videoinstruct","task":"VCGBench-Diverse","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":3,"of":6,"metrics":{"Consistency":"2.27","Contextual Understanding":"2.51","Correctness of Information":"2.13","Dense Captioning":"1.26","Detail Orientation":"2.42","Reasoning":"3.13","Spatial Understanding":"2.43","Temporal Understanding":"1.66","mean":"2.20"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-activitynet-qa","task":"Video Question Answering","dataset":"ActivityNet-QA","model":"VideoChat2","rank_in_archive_order":8,"of":36,"metrics":{"Accuracy":"49.1","Confidence score":"3.3"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-intentqa","task":"Video Question Answering","dataset":"IntentQA","model":"VideoChat2_HD_mistral","rank_in_archive_order":1,"of":6,"metrics":{"Accuarcy":"83.4","CH":"90.0","CW":"84.0","TP&TN":"77.3"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-intentqa","task":"Video Question Answering","dataset":"IntentQA","model":"VideoChat2_mistral","rank_in_archive_order":2,"of":6,"metrics":{"Accuarcy":"81.9","CH":"86.9","CW":"82.6","TP&TN":"77.0"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-mvbench","task":"Video Question Answering","dataset":"MVBench","model":"VideoChat2","rank_in_archive_order":13,"of":22,"metrics":{"Avg.":"51.9"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-next-qa","task":"Video Question Answering","dataset":"NExT-QA","model":"VideoChat2_HD_mistral","rank_in_archive_order":14,"of":47,"metrics":{"Accuracy":"79.5"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-next-qa","task":"Video Question Answering","dataset":"NExT-QA","model":"VideoChat2_mistral","rank_in_archive_order":17,"of":47,"metrics":{"Accuracy":"78.6"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-next-qa","task":"Video Question Answering","dataset":"NExT-QA","model":"VideoChat2","rank_in_archive_order":31,"of":47,"metrics":{"Accuracy":"68.6"},"uses_additional_data":false},{"leaderboard":"/sota/video-question-answering-on-tvbench","task":"Video Question Answering","dataset":"TVBench","model":"VideoChat2","rank_in_archive_order":27,"of":28,"metrics":{"Average Accuracy":"35.0"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance","task":"Video-based Generative Performance Benchmarking","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":10,"of":23,"metrics":{"Consistency":"2.84","Contextual Understanding":"3.72","Correctness of Information":"3.40","Detail Orientation":"2.91","Temporal Understanding":"2.65","mean":"3.10"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance","task":"Video-based Generative Performance Benchmarking","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":15,"of":23,"metrics":{"Consistency":"2.81","Contextual Understanding":"3.51","Correctness of Information":"3.02","Detail Orientation":"2.88","Temporal Understanding":"2.66","mean":"2.98"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-2","task":"Video-based Generative Performance Benchmarking (Consistency)","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":7,"of":18,"metrics":{"gpt-score":"2.81"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-2","task":"Video-based Generative Performance Benchmarking (Consistency)","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":10,"of":18,"metrics":{"gpt-score":"2.62"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-3","task":"Video-based Generative Performance Benchmarking (Contextual Understanding)","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":7,"of":18,"metrics":{"gpt-score":"3.64"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-3","task":"Video-based Generative Performance Benchmarking (Contextual Understanding)","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":9,"of":18,"metrics":{"gpt-score":"3.51"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-1","task":"Video-based Generative Performance Benchmarking (Correctness of Information)","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":5,"of":18,"metrics":{"gpt-score":"3.40"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-1","task":"Video-based Generative Performance Benchmarking (Correctness of Information)","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":9,"of":18,"metrics":{"gpt-score":"3.02"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-4","task":"Video-based Generative Performance Benchmarking (Detail Orientation))","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":11,"of":18,"metrics":{"gpt-score":"2.88"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-4","task":"Video-based Generative Performance Benchmarking (Detail Orientation))","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":12,"of":18,"metrics":{"gpt-score":"2.86"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-5","task":"Video-based Generative Performance Benchmarking (Temporal Understanding)","dataset":"VideoInstruct","model":"VideoChat2","rank_in_archive_order":7,"of":18,"metrics":{"gpt-score":"2.66"},"uses_additional_data":false},{"leaderboard":"/sota/video-based-generative-performance-5","task":"Video-based Generative Performance Benchmarking (Temporal Understanding)","dataset":"VideoInstruct","model":"VideoChat2_HD_mistral","rank_in_archive_order":9,"of":18,"metrics":{"gpt-score":"2.65"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-learning-on-tvqa","task":"Zero-Shot Learning","dataset":"TVQA","model":"VideoChat2","rank_in_archive_order":1,"of":1,"metrics":{"Accuracy":"40.6"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-activitynet","task":"Zero-Shot Video Question Answer","dataset":"ActivityNet-QA","model":"VideoChat2","rank_in_archive_order":14,"of":28,"metrics":{"Accuracy":"49.1","Confidence Score":"3.3"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema-1","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (fullset)","model":"VideoChat2_phi3","rank_in_archive_order":12,"of":29,"metrics":{"Accuracy":"56.7"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema-1","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (fullset)","model":"VideoChat2_HD_mistral","rank_in_archive_order":13,"of":29,"metrics":{"Accuracy":"55.8"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema-1","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (fullset)","model":"VideoChat2_mistral","rank_in_archive_order":14,"of":29,"metrics":{"Accuracy":"54.4"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (subset)","model":"VideoChat2_HD_mistral","rank_in_archive_order":6,"of":14,"metrics":{"Accuracy":"65.6"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-egoschema","task":"Zero-Shot Video Question Answer","dataset":"EgoSchema (subset)","model":"VideoChat2_mistral","rank_in_archive_order":7,"of":14,"metrics":{"Accuracy":"63.6"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msrvtt-qa","task":"Zero-Shot Video Question Answer","dataset":"MSRVTT-QA","model":"VideoChat2","rank_in_archive_order":23,"of":30,"metrics":{"Accuracy":"54.1","Confidence Score":"3.3"},"uses_additional_data":false},{"leaderboard":"/sota/zeroshot-video-question-answer-on-msvd-qa","task":"Zero-Shot Video Question Answer","dataset":"MSVD-QA","model":"VideoChat2","rank_in_archive_order":18,"of":28,"metrics":{"Accuracy":"70.0","Confidence Score":"3.9"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-next-qa","task":"Zero-Shot Video Question Answer","dataset":"NExT-QA","model":"VideoChat2","rank_in_archive_order":20,"of":27,"metrics":{"Accuracy":"61.7"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-star","task":"Zero-Shot Video Question Answer","dataset":"STAR Benchmark","model":"VideoChat2","rank_in_archive_order":1,"of":4,"metrics":{"Accuracy":"59.0"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-tvqa","task":"Zero-Shot Video Question Answer","dataset":"TVQA","model":"VideoChat_HD_mistral  (no speech)","rank_in_archive_order":4,"of":9,"metrics":{"Accuracy":"50.6"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-tvqa","task":"Zero-Shot Video Question Answer","dataset":"TVQA","model":"VideoChat_mistral  (no speech)","rank_in_archive_order":5,"of":9,"metrics":{"Accuracy":"46.4"},"uses_additional_data":false},{"leaderboard":"/sota/zero-shot-video-question-answer-on-tvqa","task":"Zero-Shot Video Question Answer","dataset":"TVQA","model":"VideoChat2 (no speech)","rank_in_archive_order":6,"of":9,"metrics":{"Accuracy":"40.6"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2311.17005","atlas_url":"https://app.syntology.ai/?focus=2311.17005","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.17005"}},"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/bytedance/tarsier","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/opengvlab/ask-anything","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/magic-research/PLLaVA","reach":{"status":"ok"}}],"summary":{"ran_fixture":2,"ran_violates":1,"ran_draft_wrong":2,"ran":2,"unverified":3},"by_repo_kind":{"official":{"samples":10,"ran":7,"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":0,"samples":[{"code_sha256_prefix":"9b4dff79d5e6102c","entry":"apply_rotary_pos_emb","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/modeling_llama.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/modeling_llama.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9b4dff79d5e6102c"}},{"code_sha256_prefix":"4cb732f513d69dfd","entry":"disabled_train","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/blip2.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/blip2.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4cb732f513d69dfd"}},{"code_sha256_prefix":"748d5a2586ca3178","entry":"get_prompt","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/conversation.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/conversation.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"748d5a2586ca3178"}},{"code_sha256_prefix":"a475c4129bb51880","entry":"get_sim","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat2/models/criterions.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat2/models/criterions.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a475c4129bb51880"}},{"code_sha256_prefix":"f64c56c357505158","entry":"interpolate_temporal_pos_embed","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat2/models/utils.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat2/models/utils.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f64c56c357505158"}},{"code_sha256_prefix":"5349fd1148712a53","entry":"load_temp_embed_with_mismatch","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat2/models/utils.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat2/models/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5349fd1148712a53"}},{"code_sha256_prefix":"b99eea6376d1e212","entry":"rotate_half","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/modeling_llama.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/modeling_llama.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b99eea6376d1e212"}},{"code_sha256_prefix":"19a98ad1eb25acff","entry":"create_eva_vit_g","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/eva_vit.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/eva_vit.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"19a98ad1eb25acff"}},{"code_sha256_prefix":"bdec7f112532de68","entry":"inflate_weight","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/eva_vit.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/eva_vit.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bdec7f112532de68"}},{"code_sha256_prefix":"0ba0ac7553b4501c","entry":"load_state_dict","repo":"opengvlab/ask-anything","repo_kind":"official","path":"video_chat/models/eva_vit.py","file_url":"https://github.com/opengvlab/ask-anything/blob/HEAD/video_chat/models/eva_vit.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0ba0ac7553b4501c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}