--- title: "MT-Video-Bench — Pan 2026" description: "Most video benchmarks for multimodal LLMs test single questions in isolation." authors: ["Yaning Pan", "Qianqian Xie", "Guohui Zhang", "Zekun Moore Wang", "Yongqian Wen", "Yuanxing Zhang", "Haoxuan Hu", "Zhiyu Pan", "Yibing Huang", "Zhidong Gan", "Yonghong Lin", "An Ping", "Shihao Li", "Yanghai Wang", "Tianhao Peng", "Jiaheng Liu"] year: 2026 venue: "Findings of the Association for Computational Linguistics: ACL 2026" source: "https://aclanthology.org/2026.findings-acl.397/" doi: "10.18653/v1/2026.findings-acl.397" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/papers/pan-2026-mt-video-bench-holistic-video/" --- [Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/pan-2026-mt-video-bench-holistic-video/) [home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2026](https://md.research.elamur.ai/free/papers/2026/page/3/) / MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues # MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues **Authors:** Yaning Pan, [Qianqian Xie](https://md.research.elamur.ai/free/authors/qianqian-xie/), Guohui Zhang, [Zekun Moore Wang](https://md.research.elamur.ai/free/authors/zekun-moore-wang/), Yongqian Wen, Yuanxing Zhang, Haoxuan Hu, Zhiyu Pan, Yibing Huang, Zhidong Gan, Yonghong Lin, An Ping et al. (16 authors) · **Year:** 2026 · **Venue:** Findings of the Association for Computational Linguistics: ACL 2026 · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2026.findings-acl.397/) · [PDF](https://aclanthology.org/2026.findings-acl.397.pdf) · [DOI 10.18653/v1/2026.findings-acl.397](https://doi.org/10.18653/v1/2026.findings-acl.397) · [OpenAlex](https://openalex.org/W4415901945) **Citations:** 0 citations (openalex-snapshot · 2026-10-14) ## Summary Most video benchmarks for multimodal LLMs test single questions in isolation. Real users, however, hold extended conversations: they follow up, change topics, and sometimes ask things the video cannot answer. MT-Video-Bench addresses this gap. It measures six skills in two groups. Perceptivity covers object reference, memory recall and content summary. Interactivity, the second group, tests whether a model can decline to answer when appropriate, cope with shifts in topic, and engage the user proactively. The dialogues were built semi-automatically. Videos were cut into scenes with PySceneDetect, and an object memory bank was used to merge related scenes. Multi-turn QA was then generated for single-scene and cross-scene settings, and humans verified the results. Scoring is checklist-based, with Gemini-2.5-Flash as the judge, and it agrees closely with human annotators. The authors tested 24 open and closed models, and all of them struggled. The best, Gemini-2.5-Pro, scored under 77%. Accuracy fell further as dialogues got longer, videos had more scenes, or clips ran longer. Early mistakes also tended to cascade through later turns. The benchmark is released on Hugging Face as NJU-LINK/MT-Video-Bench and offers a realistic testbed for conversational video assistants. ## Key points - **Scale and comparison:** the benchmark contains 1,000 dialogues and 5,887 QA pairs, averaging about 5.9 turns. Its questions and answers are longer and more lexically diverse than those of MVBench, Video-MME, MLVU or SVBench. - **Headline scores:** Gemini-2.5-Pro reached 76.95% overall. Most open-source models scored below 60%, with the Qwen3-VL series as the main exception. - **Scene and video length:** moving from 1–5 scenes to more than 20 cut Gemini-2.5-Pro's accuracy by about 13%. Performance across models also dropped clearly once videos exceeded 15 minutes. - **Dialogue history:** feeding models their own earlier answers instead of the gold history caused large error propagation. Going from 5 to 8 turns hurt every model, and InternVL3.5-8B (No Think) fell from 57.18 to 34.63. - **Input settings and judge reliability:** - More frames generally helped, but Qwen3-VL-8B saturated at 128 frames. - Most models peaked at 720p resolution. - The Gemini-2.5-Flash judge matched human ratings with a Spearman correlation of 95.25 and a Pearson correlation of 97.01. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Machine translation](https://md.research.elamur.ai/free/shelves/machine-translation/) · [Natural language processing](https://md.research.elamur.ai/free/shelves/nlp/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{pan2026mt, title = {{MT-Video-Bench: A Holistic Video Understanding Benchmark for Evaluating Multimodal LLMs in Multi-Turn Dialogues}}, author = {Pan, Yaning and Xie, Qianqian and Zhang, Guohui and Wang, Zekun Moore and Wen, Yongqian and Zhang, Yuanxing and Hu, Haoxuan and Pan, Zhiyu and Huang, Yibing and Gan, Zhidong and Lin, Yonghong and Ping, An and Li, Shihao and Wang, Yanghai and Peng, Tianhao and Liu, Jiaheng}, booktitle = {Findings of the Association for Computational Linguistics: ACL 2026}, year = {2026}, doi = {10.18653/v1/2026.findings-acl.397}, url = {https://doi.org/10.18653/v1/2026.findings-acl.397} } ``` - [← previous MQM Re-Annotation: A Technique for Collaborative Evaluation of Machine Translation](https://md.research.elamur.ai/free/papers/riley-2026-mqm-re-annotation-technique-collaborative/) - [next → MTQE.en-he: Machine Translation Quality Estimation for English-Hebrew](https://md.research.elamur.ai/free/papers/rosenbaum-2026-mtqe-en-he-machine-translation/) - [↑ back to papers of 2026](https://md.research.elamur.ai/free/papers/2026/page/3/) ## References 43 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Ge Bai, Jie Liu, Xingyuan Bu, Yancheng He, Jiaheng Liu, Zhanhui Zhou, Zhuoran Lin, Wenbo Su, Tiezheng Ge, Bo Zheng, and 1 others. 2024. Mt-bench-101: A fine-grained benchmark for evaluating large language models in multi-turn dialogues. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 7421–7454. [find](https://scholar.google.com/scholar?q=Ge+Bai%2C+Jie+Liu%2C+Xingyuan+Bu%2C+Yancheng+He%2C+Jiaheng+Liu%2C+Zhanhui+Zhou%2C+Zhuoran+Lin%2C+Wenbo+Su%2C+Tiezheng+Ge%2C+Bo+Zheng%2C+and+1+others%2E+2024%2E+Mt%2Dbench%2D101%3A+A+fine%2Dgrained+benchmark+for+evaluating+large+lang) [our card](https://md.research.elamur.ai/free/papers/bai-2024-mt-bench-101-fine-grained/) 2. Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, Wenbin Ge, Zhifang Guo, Qidong Huang, Jie Huang, Fei Huang, Binyuan Hui, Shutong Jiang, Zhaohai Li, Mingsheng Li, and 45 others. 2025a. Qwen3-vl technical report. arXiv preprint arXiv:2511.21631. [↗](https://arxiv.org/abs/2511.21631) [our card](https://md.research.elamur.ai/free/papers/bai-2025-qwen3-vl-technical-report/) 3. Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, and 1 others. 2025b. Qwen2. 5-vl technical report. arXiv preprint arXiv:2502.13923. [↗](https://arxiv.org/abs/2502.13923) [our card](https://md.research.elamur.ai/free/papers/bai-2025-qwen2-5-vl-technical/) 4. Keshigeyan Chandrasegaran, Agrim Gupta, Lea M Hadzic, Taran Kota, Jimming He, Cristóbal Eyzaguirre, Zane Durante, Manling Li, Jiajun Wu, and Li Fei-Fei. 2024. Hourvideo: 1-hour video-language understanding. Advances in Neural Information Processing Systems, 37:53168–53197. [find](https://scholar.google.com/scholar?q=Keshigeyan+Chandrasegaran%2C+Agrim+Gupta%2C+Lea+M+Hadzic%2C+Taran+Kota%2C+Jimming+He%2C+Crist%C3%B3bal+Eyzaguirre%2C+Zane+Durante%2C+Manling+Li%2C+Jiajun+Wu%2C+and+Li+Fei%2DFei%2E+2024%2E+Hourvideo%3A+1%2Dhour+video%2Dlanguage+understa) 5. Zesen Cheng, Sicong Leng, Hang Zhang, Yifei Xin, Xin Li, Guanzheng Chen, Yongxin Zhu, Wenqi Zhang, Ziyang Luo, Deli Zhao, and 1 others. 2024. Videollama 2: Advancing spatial-temporal modeling and audio understanding in video-llms. arXiv preprint arXiv:2406.07476. [↗](https://arxiv.org/abs/2406.07476) 6. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Amy Yang, Angela Fan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783. [↗](https://arxiv.org/abs/2407.21783) 7. Chaoyou Fu, Yuhan Dai, Yongdong Luo, Lei Li, Shuhuai Ren, Renrui Zhang, Zihan Wang, Chenyu Zhou, Yunhang Shen, Mengdan Zhang, and 1 others. 2025. Video-mme: The first-ever comprehensive evaluation benchmark of multi-modal llms in video analysis. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 24108– 24118. [find](https://scholar.google.com/scholar?q=Chaoyou+Fu%2C+Yuhan+Dai%2C+Yongdong+Luo%2C+Lei+Li%2C+Shuhuai+Ren%2C+Renrui+Zhang%2C+Zihan+Wang%2C+Chenyu+Zhou%2C+Yunhang+Shen%2C+Mengdan+Zhang%2C+and+1+others%2E+2025%2E+Video%2Dmme%3A+The+first%2Dever+comprehensive+evaluation+ben) [our card](https://md.research.elamur.ai/free/papers/fu-2024-video-mme-first-ever-comprehensive/) 8. Rahima Khanam and Muhammad Hussain. 2024. Yolov11: An overview of the key architectural enhancements. arXiv preprint arXiv:2410.17725. [↗](https://arxiv.org/abs/2410.17725) 9. Young-Jun Lee, Byung-Kwan Lee, Jianshu Zhang, Yechan Hwang, Byungsoo Ko, Han-Gyu Kim, Dongyu Yao, Xuankun Rong, Eojin Joo, Seung-Ho Han, and 1 others. 2025. Multiverse: A multi-turn conversation benchmark for evaluating large vision and language models. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 708–719. [find](https://scholar.google.com/scholar?q=Young%2DJun+Lee%2C+Byung%2DKwan+Lee%2C+Jianshu+Zhang%2C+Yechan+Hwang%2C+Byungsoo+Ko%2C+Han%2DGyu+Kim%2C+Dongyu+Yao%2C+Xuankun+Rong%2C+Eojin+Joo%2C+Seung%2DHo+Han%2C+and+1+others%2E+2025%2E+Multiverse%3A+A+multi%2Dturn+conversation+bench) 10. Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Peiyuan Zhang, Yanwei Li, Ziwei Liu, and 1 others. 2024a. Llava-onevision: Easy visual task transfer. arXiv preprint arXiv:2408.03326. [↗](https://arxiv.org/abs/2408.03326) [our card](https://md.research.elamur.ai/free/papers/li-2024-llava-onevision-easy-visual-task/) 11. Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Peiyuan Zhang, Yanwei Li, Ziwei Liu, and 1 others. 2024b. Llava-onevision: Easy visual task transfer. arXiv preprint arXiv:2408.03326. [↗](https://arxiv.org/abs/2408.03326) [our card](https://md.research.elamur.ai/free/papers/li-2024-llava-onevision-easy-visual-task/) 12. KunChang Li, Yinan He, Yi Wang, Yizhuo Li, Wenhai Wang, Ping Luo, Yali Wang, Limin Wang, and Yu Qiao. 2023. Videochat: Chat-centric video understanding. arXiv preprint arXiv:2305.06355. [↗](https://arxiv.org/abs/2305.06355) [our card](https://md.research.elamur.ai/free/papers/li-2023-videochat-chat-centric-video-understanding/) 13. Kunchang Li, Yali Wang, Yinan He, Yizhuo Li, Yi Wang, Yi Liu, Zun Wang, Jilan Xu, Guo Chen, Ping Luo, and 1 others. 2024c. Mvbench: A comprehensive multi-modal video understanding benchmark. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 22195–22206. [find](https://scholar.google.com/scholar?q=Kunchang+Li%2C+Yali+Wang%2C+Yinan+He%2C+Yizhuo+Li%2C+Yi+Wang%2C+Yi+Liu%2C+Zun+Wang%2C+Jilan+Xu%2C+Guo+Chen%2C+Ping+Luo%2C+and+1+others%2E+2024c%2E+Mvbench%3A+A+comprehensive+multi%2Dmodal+video+understanding+benchmark%2E+In+Procee) [our card](https://md.research.elamur.ai/free/papers/li-2023-mvbench-comprehensive-multi-modal-video/) 14. Xinhao Li, Yi Wang, Jiashuo Yu, Xiangyu Zeng, Yuhan Zhu, Haian Huang, Jianfei Gao, Kunchang Li, Yinan He, Chenting Wang, and 1 others. 2024d. Videochat-flash: Hierarchical compression for long-context video modeling. arXiv preprint arXiv:2501.00574. [↗](https://arxiv.org/abs/2501.00574) 15. Xinhao Li, Yi Wang, Jiashuo Yu, Xiangyu Zeng, Yuhan Zhu, Haian Huang, Jianfei Gao, Kunchang Li, Yinan He, Chenting Wang, and 1 others. 2024e. Videochat-flash: Hierarchical compression for long-context video modeling. arXiv preprint arXiv:2501.00574. [↗](https://arxiv.org/abs/2501.00574) 16. Bingchen Liu, Ehsan Akhgari, Alexander Visheratin, Aleks Kamko, Linmiao Xu, Shivam Shrirao, Chase Lambert, Joao Souza, Suhail Doshi, and Daiqing Li. 2024. Playground v3: Improving text-to-image alignment with deep-fusion large language models. arXiv preprint arXiv:2409.10695. [↗](https://arxiv.org/abs/2409.10695) 17. Fan Ma, Xiaojie Jin, Heng Wang, Yuchen Xian, Jiashi Feng, and Yi Yang. 2024. Vista-llama: Reducing hallucination in video language models via equal distance to visual tokens. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 13151–13160. [find](https://scholar.google.com/scholar?q=Fan+Ma%2C+Xiaojie+Jin%2C+Heng+Wang%2C+Yuchen+Xian%2C+Jiashi+Feng%2C+and+Yi+Yang%2E+2024%2E+Vista%2Dllama%3A+Reducing+hallucination+in+video+language+models+via+equal+distance+to+visual+tokens%2E+In+Proceedings+of+the+IEE) 18. Muhammad Maaz, Hanoona Rasheed, Salman Khan, and Fahad Shahbaz Khan. 2023. Video-chatgpt: Towards detailed video understanding via large vision and language models. arXiv preprint arXiv:2306.05424. [↗](https://arxiv.org/abs/2306.05424) [our card](https://md.research.elamur.ai/free/papers/maaz-2023-video-chatgpt-towards-detailed-video/) 19. Ruchit Rawal, Khalid Saifullah, Miquel Farré, Ronen Basri, David Jacobs, Gowthami Somepalli, and Tom Goldstein. 2024. Cinepile: A long video question answering dataset and benchmark. arXiv preprint arXiv:2405.08813. [↗](https://arxiv.org/abs/2405.08813) 20. Enxin Song, Wenhao Chai, Weili Xu, Jianwen Xie, Yuxuan Liu, and Gaoang Wang. 2025. Video-mmlu: A massive multi-discipline lecture understanding benchmark. arXiv preprint arXiv:2504.14693. [↗](https://arxiv.org/abs/2504.14693) 21. ByteDance Seed Team. 2025a. Doubao-seed-1.6. [https://www.volcengine.com/product/doubao](https://www.volcengine.com/product/doubao). [↗](https://www.volcengine.com/product/doubao) 22. Gemini Team. 2025b. Build rich, interactive web apps with an updated gemini 2.5 pro. [https://deepmind.google/models/gemini/](https://deepmind.google/models/gemini/). [↗](https://deepmind.google/models/gemini/) 23. Gemini Team. 2025c. A new era of intelligence with gemini 3. [https://deepmind.google/models/gemini/](https://deepmind.google/models/gemini/). [↗](https://deepmind.google/models/gemini/) 24. Haochen Wang, Yucheng Zhao, Tiancai Wang, Haoqiang Fan, Xiangyu Zhang, and Zhaoxiang Zhang. 2025a. Ross3d: Reconstructive visual instruction tuning with 3d-awareness. arXiv preprint arXiv:2504.01901. [↗](https://arxiv.org/abs/2504.01901) 25. Haochen Wang, Anlin Zheng, Yucheng Zhao, Tiancai Wang, Zheng Ge, Xiangyu Zhang, and Zhaoxiang Zhang. 2024a. Reconstructive visual instruction tuning. arXiv preprint arXiv:2410.09575. [↗](https://arxiv.org/abs/2410.09575) 26. Weihan Wang, Zehai He, Wenyi Hong, Yean Cheng, Xiaohan Zhang, Ji Qi, Xiaotao Gu, Shiyu Huang, Bin Xu, Yuxiao Dong, and 1 others. 2024b. Lvbench: An extreme long video understanding benchmark. arXiv preprint arXiv:2406.08035. [↗](https://arxiv.org/abs/2406.08035) 27. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025b. Internvl3. 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265. [↗](https://arxiv.org/abs/2508.18265) [our card](https://md.research.elamur.ai/free/papers/wang-2025-internvl3-5-advancing-open/) 28. Weiyun Wang, Zhangwei Gao, Lixin Gu, Hengjun Pu, Long Cui, Xingguang Wei, Zhaoyang Liu, Linglin Jing, Shenglong Ye, Jie Shao, and 1 others. 2025c. Internvl3. 5: Advancing open-source multimodal models in versatility, reasoning, and efficiency. arXiv preprint arXiv:2508.18265. [↗](https://arxiv.org/abs/2508.18265) [our card](https://md.research.elamur.ai/free/papers/wang-2025-internvl3-5-advancing-open/) 29. Yi Wang, Xinhao Li, Ziang Yan, Yinan He, Jiashuo Yu, Xiangyu Zeng, Chenting Wang, Changlian Ma, Haian Huang, Jianfei Gao, and 1 others. 2025d. Internvideo2. 5: Empowering video mllms with long and rich context modeling. arXiv preprint arXiv:2501.12386. [↗](https://arxiv.org/abs/2501.12386) 30. Yi Wang, Xinhao Li, Ziang Yan, Yinan He, Jiashuo Yu, Xiangyu Zeng, Chenting Wang, Changlian Ma, Haian Huang, Jianfei Gao, and 1 others. 2025e. Internvideo2. 5: Empowering video mllms with long and rich context modeling. arXiv preprint arXiv:2501.12386. [↗](https://arxiv.org/abs/2501.12386) 31. Yu Wang, Zeyuan Zhang, Julian McAuley, and Zexue He. 2024c. Lvchat: Facilitating long video comprehension. arXiv preprint arXiv:2402.12079. [↗](https://arxiv.org/abs/2402.12079) 32. Haoning Wu, Dongxu Li, Bei Chen, and Junnan Li. 2024. Longvideobench: A benchmark for long-context interleaved video-language understanding. Advances in Neural Information Processing Systems, 37:28828–28857. [find](https://scholar.google.com/scholar?q=Haoning+Wu%2C+Dongxu+Li%2C+Bei+Chen%2C+and+Junnan+Li%2E+2024%2E+Longvideobench%3A+A+benchmark+for+long%2Dcontext+interleaved+video%2Dlanguage+understanding%2E+Advances+in+Neural+Information+Processing+Systems%2C+37%3A28828) 33. Zhenyu Yang, Yuhang Hu, Zemin Du, Dizhan Xue, Shengsheng Qian, Jiahong Wu, Fan Yang, Weiming Dong, and Changsheng Xu. 2025. Svbench: A benchmark with temporal multi-turn dialogues for streaming video understanding. arXiv preprint arXiv:2502.10810. [↗](https://arxiv.org/abs/2502.10810) 34. Yuan Yao, Tianyu Yu, Ao Zhang, Chongyi Wang, Junbo Cui, Hongji Zhu, Tianchi Cai, Haoyu Li, Weilin Zhao, Zhihui He, and 1 others. 2024. Minicpm-v: A gpt-4v level mllm on your phone. arXiv preprint arXiv:2408.01800. [↗](https://arxiv.org/abs/2408.01800) [our card](https://md.research.elamur.ai/free/papers/yao-2024-minicpm-v-gpt-4v/) 35. Seonghyeon Ye, Doyoung Kim, Sungdong Kim, Hyeonbin Hwang, Seungone Kim, Yongrae Jo, James Thorne, Juho Kim, and Minjoon Seo. 2023. Flask: Fine-grained language model evaluation based on alignment skill sets. arXiv preprint arXiv:2307.10928. [↗](https://arxiv.org/abs/2307.10928) 36. Jiashuo Yu, Yue Wu, Meng Chu, Zhifei Ren, Zizheng Huang, Pei Chu, Ruijie Zhang, Yinan He, Qirui Li, Songze Li, and 1 others. 2025. Vrbench: A benchmark for multi-step reasoning in long narrative videos. arXiv preprint arXiv:2506.10857. [↗](https://arxiv.org/abs/2506.10857) 37. Boqiang Zhang, Kehan Li, Zesen Cheng, Zhiqiang Hu, Yuqian Yuan, Guanzheng Chen, Sicong Leng, Yuming Jiang, Hang Zhang, Xin Li, and 1 others. 2025a. Videollama 3: Frontier multimodal foundation models for image and video understanding. arXiv preprint arXiv:2501.13106. [↗](https://arxiv.org/abs/2501.13106) 38. Haoji Zhang, Yiqin Wang, Yansong Tang, Yong Liu, Jiashi Feng, Jifeng Dai, and Xiaojie Jin. 2024a. Flash-vstream: Memory-based real-time understanding for long video streams. arXiv preprint arXiv:2406.08085. [↗](https://arxiv.org/abs/2406.08085) 39. Yuanhan Zhang, Bo Li, Haotian Liu, Yong Jae Lee, Liangke Gui, Di Fu, Jiashi Feng, Ziwei Liu, and Chunyuan Li. 2024b. Llava-next: A strong zero-shot video understanding model. [https://llava-vl.github.io/blog/2024-04-30-llava-next/](https://llava-vl.github.io/blog/2024-04-30-llava-next/). [↗](https://llava-vl.github.io/blog/2024-04-30-llava-next/) 40. Yuanhan Zhang, Jinming Wu, Wei Li, Bo Li, Zejun Ma, Ziwei Liu, and Chunyuan Li. 2024c. Video instruction tuning with synthetic data. arXiv preprint arXiv:2410.02713. [↗](https://arxiv.org/abs/2410.02713) 41. Yuanhan Zhang, Jinming Wu, Wei Li, Bo Li, Zejun Ma, Ziwei Liu, and Chunyuan Li. 2025b. Llava-video: Video instruction tuning with synthetic data. Transactions on Machine Learning Research. [↗](https://doi.org/10.48550/arxiv.2410.02713) 42. Junjie Zhou, Yan Shu, Bo Zhao, Boya Wu, Zhengyang Liang, Shitao Xiao, Minghao Qin, Xi Yang, Yongping Xiong, Bo Zhang, and 1 others. 2025. Mlvu: Benchmarking multi-task long video understanding. In Proceedings of the Computer Vision and Pattern Recognition Conference, pages 13691–13701. [find](https://scholar.google.com/scholar?q=Junjie+Zhou%2C+Yan+Shu%2C+Bo+Zhao%2C+Boya+Wu%2C+Zhengyang+Liang%2C+Shitao+Xiao%2C+Minghao+Qin%2C+Xi+Yang%2C+Yongping+Xiong%2C+Bo+Zhang%2C+and+1+others%2E+2025%2E+Mlvu%3A+Benchmarking+multi%2Dtask+long+video+understanding%2E+In+Pro) 43. Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, and Mohamed Elhoseiny. 2023. Minigpt-4: Enhancing vision-language understanding with advanced large language models. arXiv preprint arXiv:2304.10592. [↗](https://arxiv.org/abs/2304.10592) [our card](https://md.research.elamur.ai/free/papers/zhu-2023-minigpt-4-enhancing-vision-language/)