The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index.
MLVU: Benchmarking Multi-task Long Video Understanding
Open accessAbstract
The evaluation of Long Video Understanding (LVU) performance poses an important but challenging research problem. Despite previous efforts, the existing video understanding benchmarks are severely constrained by several issues, especially the insufficient lengths of videos, a lack of diversity in video types and evaluation tasks, and the inappropriateness for evaluating LVU performances. To address the above problems, we propose a new benchmark called MLVU (Multitask Long Video Understanding Benchmark) for the comprehensive and in-depth evaluation of LVU. MLVU presents the following critical values: 1) The substantial and flexible extension of video lengths, which enables the benchmark to evaluate LVU performance across a wide range of durations. 2) The inclusion of various video genres, such as movies, surveillance, egocentric videos, and cartoons, reflects the models’ LVU performances in different scenarios. 3) The development of diversified evaluation tasks, which enables a comprehensive examination of MLLMs’ key abilities in long-video understanding. The empirical study with 23 latest MLLMs reveals significant room for improvement in today’s technique, as all existing methods struggle with most of the evaluation tasks and exhibit severe performance degradation when handling longer videos. Additionally, it suggests that factors such as context length, image-understanding ability, and the choice of LLM backbone can play critical roles in future advancements. We anticipate that MLVU will advance the research of LVU by providing a comprehensive and in-depth analysis of MLLMs. The code and dataset can be accessed from https://github.com/JUNJIE99/MLVU.
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Most existing LVQA benchmarks [2, 11, 50, 62] mainly focus on visual understanding, such as action recognition and spatial/temporal perception, leaving a gap in evaluating higher-level cognitive abilities that rely on long-term memory and are crucial for real-world agents, such as understanding persons, extracting general knowledge, and performing cross-modal reasoning.”
From Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memory (2025) · cites this paper as [62]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 8
- [6][6] Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners. Advances in neural information processing systems, 33:1877-1901, 2020. 1 -
- [15][15] Kristen Grauman, Andrew Westbury, Eugene Byrne, Zachary Chavis, Antonino Furnari, Rohit Girdhar, Jackson Hamburger, Hao Jiang, Miao Liu, Xingyu Liu, et al. Ego4d: Around the world in 3,000 hours of egocentric video. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 18995-19012, 2022. 1, 3, 4 -
- [34][34] Karttikeya Mangalam, Raiymbek Akshulakov, and Jitendra Malik. Egoschema: A diagnostic benchmark for very longform video language understanding. Advances in Neural Information Processing Systems, 36, 2023. 1, 2, 3 -
- [48][48] Bo Wu, Shoubin Yu, Zhenfang Chen, Joshua B Tenenbaum, and Chuang Gan. Star: A benchmark for situated reasoning in real-world videos. In Thirty-fifth conference on neural information processing systems datasets and benchmarks track (Round 2), 2021. 3 -
- [52][52] Jun Xu, Tao Mei, Ting Yao, and Yong Rui. Msr-vtt: A large video description dataset for bridging video and language. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5288-5296, 2016. 1, 2, 3 -
- [55][55] Kexin Yi, Chuang Gan, Yunzhu Li, Pushmeet Kohli, Jiajun Wu, Antonio Torralba, and Joshua B Tenenbaum. Clevrer: Collision events for video representation and reasoning. In International Conference on Learning Representations, 2019. 4 -
- [56][56] Zhou Yu, Dejing Xu, Jun Yu, Ting Yu, Zhou Zhao, Yueting Zhuang, and Dacheng Tao. Activitynet-qa: A dataset for understanding complex web videos via question answering. In Proceedings of the AAAI Conference on Artificial Intelligence, pages 9127-9134, 2019. 3 -
- [63][63] Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, and Mohamed Elhoseiny. Minigpt-4: Enhancing visionlanguage understanding with advanced large language models. In The Twelfth International Conference on Learning Representations, 2023. 2
Cited by
Sorted by citations
Items 201–300 of 454
Divide, then Ground: Adapting Frame Selection to Query Types for Long-Form Video UnderstandingPreprint
Li, J., Li, B., Li, J., & Lu, Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.04000 · Cited by 8
MVU-Eval: Towards Multi-Video Understanding Evaluation for Multimodal LLMsPreprint
Peng, T., Wang, H., Zhang, Y., Wang, Z., Wang, Z., Chang, G., Yang, J., Li, S., Wang, Y., Wang, X., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.07250 · Cited by 8
Scaling the Long Video Understanding of Multimodal Large Language Models via Visual Memory MechanismPreprint
Chen, T., Zhang, K., Wu, Q., Chen, X., Chang, C., Sun, X., Zhou, Y., & Ji, R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.29252 · Cited by 8
Black Swan: Abductive and Defeasible Video Reasoning in Unpredictable Events
Chinchure, A., Ravi, S., Ng, R. T., Shwartz, V., Li, B., & Sigal, L. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.02254 · Cited by 7
Bidirectional Likelihood Estimation with Multi-Modal Large Language Models for Text-Video Retrieval
Ko, D., Lee, J. S., Choi, M., Meng, Z., & Kim, H. J. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02067 · Cited by 7
EgoExo-Con: Exploring View-Invariant Video Temporal Understanding
Jung, M., Xiao, J., Kim, J., Zhang, B.-T., & Yao, A. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37225-3_20 · Cited by 7
Self-citationFocus: A Streaming Concentration Architecture for Efficient Vision-Language Models
Wei, C., Guo, C., Zhang, J., Shan, H., Xu, Y., Zhang, Z., Liu, Y., Wang, Q., Zhou, C., Li, H. H., et al. (2026) · IEEE International Symposium on High-Performance Computer Architecture (HPCA) · doi:10.1109/hpca68181.2026.11408525 · Cited by 8
AdaTP: Attention-Debiased Token Pruning for Video Large Language Models
Sun, F., Shen, L., Chen, H., Zhao, S., Han, J., & Ding, G. (2025) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2025.findings-emnlp.174 · Cited by 7
TinyLLaVA-Video: Towards Smaller LMMs for Video Understanding with Group ResamplerPreprint
Zhang, X., Weng, X., Yue, Y., Fan, Z., Wu, W., & Huang, L. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2501.15513 · Cited by 7
IV-Bench: A Benchmark for Image-Grounded Video Perception and Reasoning in Multimodal LLMsPreprint
Ma, D., Zhang, Y., Ren, J., Guo, J., Yao, Y., Wei, Z., Yang, Z., Peng, Z., Feng, B., Ma, J., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2504.15415 · Cited by 8
FLoC: Facility Location-Based Efficient Visual Token Compression for Long Video UnderstandingPreprint
Cho, J., Lee, J., Hayat, M., Hwang, K., Porikli, F., & Choi, S. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.00141 · Cited by 7
VideoLLM Benchmarks and Evaluation: A SurveyPreprint
Kumar, Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.03829 · Cited by 7
IF-VidCap: Can Video Caption Models Follow Instructions?Preprint
Li, S., Zhang, Y., Wu, J., Lei, Z., He, Y., Wen, R., Liao, C., Jiang, C., Ping, A., Gao, S., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2510.18726 · Cited by 7
Self-citationAURA: A Fine-Grained Benchmark and Decomposed Metric for Audio-Visual ReasoningPreprint
Galougah, S. S., Raj, R., Chowdhury, S., Nag, S., & Ramani Duraiswami, B. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.07470 · Cited by 7
VCapsBench: A Large-scale Fine-grained Benchmark for Video Caption Quality EvaluationPreprint
Zhang, S.-X., Wang, H., Huang, D., Li, X., Zhu, X., & Yin, X.-C. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.23484 · Cited by 7
MA-Bench: Towards Fine-grained Micro-Action UnderstandingPreprint
Li, K., Gu, J., Wang, F., Wu, Z., Fan, H., & Guo, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.26586 · Cited by 7
AdaCodec: A Predictive Visual Code for Video MLLMsPreprint
Hou, H., Huang, Z., Liang, Z., Si, Q., Li, C., Dong, S., Shao, K., Li, R., Wang, D., Duan, N., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.02569 · Cited by 7
Self-citationWatch, Remember, Reason: Human-View Video Understanding with MLLMsPreprint
Meng, J., Tan, Y., Xu, Q., Gao, K., Liu, W., Li, Y., Li, J., Kong, L., Wang, H., Zhou, Q., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.07433 · Cited by 7
EgoMonth: A Month-Level Egocentric Video Benchmark for Long-Term Spatiotemporal MemoryPreprint
Chen, W., Jiaxin, H., Xie, T., Li, Y., Qian, Y., Xu, B., Tang, Z., Wang, S., Yu, M., Li, D., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.13113 · Cited by 7
LongVidSearch: An Agentic Benchmark for Multi-hop Evidence Retrieval Planning in Long Videos
Yu, R., Duan, C., Liang, H., Sun, H., & Pei, P. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3838652 · Cited by 6
Towards general auditory intelligence for machine listening and speaking
Wang, S., Jin, Z., Tang, C., Li, Q., Li, B., Chen, C., Hu, Y., Yu, W., Li, Y., Zhuang, J., et al. (2026) · Nature Machine Intelligence · doi:10.1038/s42256-026-01281-1 · Cited by 6
Breaking the Encoder Barrier for Seamless Video-Language Understanding
Li, H., Zhang, Y., Guo, L., Yue, X., & Liu, J. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02151 · Cited by 6
Self-alignment of Large Video Language Models with Refined Regularized Preference Optimization
Sarkar, P., & Etemad, A. (2025) · neural information processing systems · doi:10.52202/085713-1420 · Cited by 6
ExAct: A Video-Language Benchmark for Expert Action Analysis
Han, Y., Pan, Y., He, F., Liu, X., Zhang, B., Oguntola, O., & Bertasius, G. (2025) · neural information processing systems · doi:10.52202/085713-0246 · Cited by 6
Self-citationMobile-VideoGPT: Fast and Accurate Model for Mobile Video UnderstandingPreprint
Shaker, A., Maaz, M., Gou, C., Rezatofighi, H., Khan, S., & Khan, F. S. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.21782 · Cited by 6
On Path to Multimodal Generalist: General-Level and General-BenchPreprint
Fei, H., Zhou, Y., Li, J., Li, X., Xu, Q., Li, B., Wu, S., Wang, Y., Zhou, J., Meng, J., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.04620 · Cited by 6
Self-citationVCRBench: Exploring Long-form Causal Reasoning Capabilities of Large Video Language ModelsPreprint
Sarkar, P., & Etemad, A. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.08455 · Cited by 6
A Benchmark for Omni-Modal Reasoning in Long VideosPreprint
Kurpath, M. I., Kaithakkodan, J. M., Zhou, J., Mullappilly, S. S., Almansoori, M., Ahsan, N., Kalmakhanbet, B., Shikhar, S., Lalla, R., Lahoud, J., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.16978 · Cited by 6
Self-citationChung, S., Kim, Y. M., Chee, Y., & Ro, Y. M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2601.21181 · Cited by 6
MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-AttentionPreprint
Wang, P., Tan, C., Zhou, S., (36889), W. H., Zhou, Q., Huang, Z., Ye, Z., Cheng, J., Qian, X., Chen, Y., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.07639 · Cited by 6
VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding
Zhang, B., Li, K., Cheng, Z., Hu, Z., Yuan, Y., Chen, G., Leng, S., Jiang, Y., Zhang, H., Li, X., et al. (2025) · arXiv · doi:10.48550/arxiv.2501.13106 · Cited by 5
Self-citationMultimodal Fusion and Coherence Modeling for Video Topic Segmentation
Yu, H., Deng, C., Zhang, Q., Liu, J., Chen, Q., & Wang, W. (2025) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2025.findings-acl.904 · Cited by 5
VF-Eval: Evaluating Multimodal LLMs for Generating Feedback on AIGC Videos
Song, T., Hu, T., Gan, G., & Zhao, Y. (2025) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2025.acl-long.1027 · Cited by 5
PAVE: Patching and Adapting Video Large Language Models
Liu, Z., Li, Y., Nguyen, K. D., Zhong, Y., & Li, Y. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.00314 · Cited by 5
Self-citationFlow4Agent: Long-form Video Understanding via Motion Prior from Optical Flow
Liu, R., Sun, S., Tang, H., Gao, W., & Li, G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02211 · Cited by 5
Cao, Z., He, Y., Liu, A., Xie, J., Chen, F., & Wang, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755873 · Cited by 5
Sparrow: Data-Efficient Video-LLM With Text-to-Image Augmentation
Yin, S., Fu, C., Zhao, S., Ge, C., Yang, Y., Dai, Y., Luo, Y., Xu, T., Shan, C., & Chen, E. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3673542 · Cited by 5
Static or Dynamic: Towards Query-Adaptive Token Selection for Video Question Answering
Shi, Y., Long, Q., & Wang, W. (2025) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2025.emnlp-main.545 · Cited by 5
Infinite Video UnderstandingPreprint
Zhang, D., Chen, X., Luo, J., Jia, M., Sun, C., Ren, R., Liu, J., Sun, H., & Li, X. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.09068 · Cited by 5
Zou, J., Xia, H., Ye, Z., Zhang, S., Lai, C., Ordóñez, V., Shen, W., & Chen, H. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.12908 · Cited by 5
LaVi: Efficient Large Vision-Language Models via Internal Feature ModulationPreprint
Yue, T., Guo, L., Tang, Y., Zhao, Z., Zhu, X., Huang, H., & Liu, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.16691 · Cited by 5
MSJoE: Jointly Evolving MLLM and Sampler for Efficient Long-Form Video UnderstandingPreprint
Tan, W., Yu, X., Li, J., Chen, Y., Ju, J., Luo, Z., Song, R., & Luan, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.22932 · Cited by 5
GIFT: Global Irreplaceability Frame Targeting for Efficient Video UnderstandingPreprint
Ma, J., Zhou, S., Li, G., Gao, X., Cao, Y., Zeng, H., Yan, Y., Wang, Z., Song, J., Zheng, B., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.25072 · Cited by 5
Diagnosing Long-Video Quantitative Reasoning in Multimodal LLMs via Enumeration and CountingPreprint
Tsuchiya, F., Miyanishi, T., Yasuki, S., Ukai, M., Inoue, N., Kurita, S., Iwasawa, Y., & Matsuo, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.29943 · Cited by 5
SVAgent: Storyline-Guided Long Video Understanding via Cross-Modal Multi-Agent CollaborationPreprint
Yang, Z., Yang, Z., Zhan, S., Yue, T., Pang, W., & Yuan, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.05079 · Cited by 5
Video-Zero: Self-Evolution Video UnderstandingPreprint
Zhang, R., Ji, D., Zhu, L., Liu, X., Meng, Y., Chu, R., & Yang, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.14733 · Cited by 5
An Efficient Streaming Video Understanding Framework with Agentic ControlPreprint
Liu, J., Huang, J., Jia, Z., Li, J., Zhang, X., Guo, Z., Li, B., Zeng, W., Lu, Y., & Jin, X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.17921 · Cited by 5
Zhang, K., Fang, C., Chen, T. X., Song, B., Shen, Y., Zhou, Y., & Ji, R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.24187 · Cited by 5
FOLIO: Focused Semantic Memory for Streaming Video UnderstandingPreprint
Fan, H., Parikh, D., Ramachandran, A., Gobriel, S., Jain, N., Kannan, R., & Prasanna, V. K. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.13298 · Cited by 5
Training-Free Uncertainty Guidance for Complex Visual Tasks with MLLMs
Kim, S., Xiao, R., Alaniz, S., Xian, Y., & Akata, Z. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37020-4_33 · Cited by 4
MemoryCard: Topic-Aware Multi-Modal Clue Compression for Long-Video Question Answering
Yang, Q., Huang, P., Li, X., Liu, Z., Yan, Y., Gu, Y., Yu, G., Li, G., & Sun, M. (2026) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/2026.emnlp-main.553 · Cited by 4
Self-citationChen, H., Wang, X., Li, G., Zhou, Y., Chen, Y., Jia, J., & Zhu, W. (2026) · International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR) · doi:10.1145/3805712.3809757 · Cited by 4
Dynamic Hub-and-Spoke Memory for Streaming Video Understanding
Jiang, X., Zhao, L., Xiao, X., Zhang, Y., Wang, J., Ma, C., Li, H., Wang, Y., Gong, Y., & CAMPS, O. (2026) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/2026.findings-emnlp.2216 · Cited by 4
Zhao, J., Sun, B., Chen, X., & Wei, X. (2026) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v40i16.38317 · Cited by 4
VEU-Bench: Towards Comprehensive Understanding of Video Editing
Li, B., Wu, Y., Lu, Y., Yu, J., Tang, L., Cao, J., Zhu, W., Sun, Y., Wu, J., & Zhu, W. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01276 · Cited by 4
Exploring the Design Space of Visual Context Representation in Video MLLMsPreprint
Du, Y., Huo, Y., Zhou, K., Zhao, Z., Lu, H., Huang, H. P., Zhao, W. X., Wang, B., Chen, W., & Wen, J.-R. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2410.13694 · Cited by 4
HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric ScenesPreprint
Li, K., Shen, H., Shi, H., Hou, R., Chang, H., Huang, J., Jia, C., Wang, W., Wu, Y., Jiang, D., et al. (2025) · arXiv (Cornell University) · Cited by 4
Sparsity Forcing: Reinforcing Token Sparsity of MLLMsPreprint
Chen, F., He, Y., Lin, L., Gou, C., Liu, J., Zhuang, B., & Wu, Q. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2504.18579 · Cited by 4
A Comprehensive Survey on Multimodal RAG: All Combinations of Modalities as Input and OutputPreprint
Zhang, R., Liu, C., Su, Y., Li, R., Huang, X., Li, X., & Yu, P. S. (2025) · doi:10.36227/techrxiv.176341513.38473003/v2 · Cited by 4
Chen, B., Wang, Z., Yue, Z., Yan, K., Yu, C., Huang, Y., Liu, Z., Wen, Y., Chen, X., Liu, Y., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.19524 · Cited by 4
Self-citationH2VU-Benchmark: A Comprehensive Benchmark for Hierarchical Holistic Video UnderstandingPreprint
Wu, Q., Zheng, Q., Zhang, Y., Xie, J., Luo, J., Wang, K., Liu, P., Xie, Q., Zhen, R., Yang, Z., et al. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.24008 · Cited by 4
TUNA: Comprehensive Fine-grained Temporal Understanding Evaluation on Dense Dynamic VideosPreprint
Kong, F., Zhang, J., Zhang, H., Shi, F., Wang, D., Yu, L., Ji, X., Tian, Y., Victoria, W., & Zhang, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.20124 · Cited by 4
MLLM-based Discovery of Intrinsic Coordinates and Governing Equations from High-Dimensional DataPreprint
Li, R., Lu, Y., Tang, S., Qi, B., & Ouyang, W. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.11940 · Cited by 4
Focus-Scan-Refine: From Human Visual Perception to Efficient Visual Token PruningPreprint
Tong, E., Bai, Y., Zhu, Y., Jiang, J., & Liu, X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.05809 · Cited by 4
EgoSound: Benchmarking Sound Understanding in Egocentric VideosPreprint
Zhu, B., Fu, Y., Dong, Q., Sun, G., Qian, T., Wu, Y., Paudel, D. P., Xue, X., & Fu, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.14122 · Cited by 4
EasyVideoR1: Easier RL for Video UnderstandingPreprint
Qin, C., Yang, C., Si, Q., Gu, N., Yao, D., Lin, Z., Fu, P., Duan, N., & Wang, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.16893 · Cited by 4
Harnessing Streaming Video in the WildPreprint
Yao, D., Gu, S., Si, Q., Zhou, J., Yang, C., Qin, C., Gu, N., Lin, Z., Wang, W., Duan, N., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.08615 · Cited by 4
Self-citationMulti-Modal Retrieval Augmented Visual Understanding and Generation
Xue, Z. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3756866 · Cited by 3
HFS: Holistic Query-Aware Frame Selection for Efficient Video Understanding
Yang, Y., Li, Y., Qi, D., Yang, L., Wang, T., Zhang, W., Li, S., & Lam, K.-M. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.48550/arxiv.2512.11534 · Cited by 3
VCapsBench: A Large-scale Fine-grained Benchmark for Video Caption Quality Evaluation
Zhang, S.-X., Wang, H., Huang, D., Li, X., Zhu, X., & Yin, X.-C. (2026) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v40i15.38269 · Cited by 3
SVCBench: A Streaming Video Counting Benchmark for Spatial-Temporal State Maintenance
Liu, P., Shi, Z., Hao, H., Fu, Q., Bi, X., Zhang, S., Hu, X., Wang, Z., Huang, L., & Liu, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37044-0_29 · Cited by 3
Video Understanding: From Geometry and Semantics to Unified Models
An, Z., Li, Z., Ye, M., Qiao, F., Li, J., Wu, Z., Thengane, V., Li, C., Li, L., Gool, L. V., et al. (2026) · Machine Intelligence Research · doi:10.1007/s11633-026-1656-7 · Cited by 3
Video-Oasis: Rethinking Evaluation of Video Understanding
Lim, G., Park, S., Lee, J., Lee, I., Kim, T., Wee, D., Shim, M., & Choi, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37095-2_33 · Cited by 3
Zeng, K., Zhong, G., Cheng, J., Yuan, J., & Li, Z. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.00221 · Cited by 3
EgoSAT: A Comprehensive Benchmark of Egocentric Streaming Interaction Understanding
Lei, Y., Li, J., Zhang, Y., Hua, J., Li, Y., & Liu, M. Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37624-4_34 · Cited by 3
QCA: Query- and Content-Aware Keyframe Selection for Long Video Understanding
Peng, J., Song, B., Li, J., Li, H., Zhou, Y., Ji, R., & Tian, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37321-2_2 · Cited by 3
Fine-Tuning Large Multimodal Models for Fitness Action Quality Assessment
Dibenedetto, G., Musacchio, E., Polignano, M., & Lops, P. (2025) · ACM Conference on User Modeling, Adaptation and Personalization (UMAP) · doi:10.1145/3708319.3733684 · Cited by 3
MESH - Understanding Videos Like Human: Measuring Hallucinations in Large Video Models
Yang, G., Chen, Z., Wong, M. H., Lei, H., Chen, Y., Li, Z., Zhou, K., & Cheng, J. S.-C. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755626 · Cited by 3
CoViPAL: Layer-wise Contextualized Visual Token Pruning for Large Vision-Language Models
Tang, Z., Ma, Z., Wang, S., Li, Z., Zhang, L., Zhao, H., Li, Y., & Wang, Q. (2025) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2025.findings-emnlp.1127 · Cited by 3
MoviePuzzle: Visual Narrative Reasoning through Multimodal Order LearningPreprint
Wang, J., Wang, Y., Zhao, D., & Zheng, Z. (2023) · arXiv (Cornell University) · doi:10.48550/arxiv.2306.02252 · Cited by 3
Chen, X., Li, Y., Shi, H., Hu, B., Luo, W., Wang, Y., & Zhang, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2504.17821 · Cited by 3
Bonnetto, A., Qi, H., Leong, F., Tashkovska, M., Rad, M., Shokur, S., Hummel, F. C., Micera, S., Pollefeys, M., & Mathis, A. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.01608 · Cited by 3
Enhancing Long Video Question Answering with Scene-Localized Frame GroupingPreprint
Yang, X., Zhang, W., Jin, H., Liu, L., Xu, H., Nie, Y., Yu, F., & Ma, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.03009 · Cited by 3
Self-citationEfficient Motion-Aware Video MLLMPreprint
Zhao, Z., Huo, Y., Yue, T., Guo, L., Lu, H., Wang, B., Chen, W., & Liu, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.13016 · Cited by 3
Chen, T., Ju, S., Wu, Q., Fang, C., Zhang, K., Peng, J., Li, H., Zhou, Y., & Ji, R. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.08410 · Cited by 3
Ju, S., Song, B., Chen, T., Zhang, J., Wu, Q., Chang, C., Wang, H., Zhou, Y., & Ji, R. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.22911 · Cited by 3
Li, A., Numan, N., & Steed, A. J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.16481 · Cited by 3
Benchmarking Visual State Tracking in Multimodal Video UnderstandingPreprint
Yu, S., Ma, N., Huang, P., Lee, H., Yang, S., Choi, J. S., Brown, E., Michel, O., Zheng, B., Shin, J., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.03920 · Cited by 3
Perception Before Reasoning: Dynamic Latent Reasoning for Video Understanding and Question AnsweringPreprint
Xia, H., Xiao, Z., Zou, J., Ordóñez, V., & Chen, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.04124 · Cited by 3
VideoGAIA: A Benchmark for General AI Assistants on Agentic Video UnderstandingPreprint
Zhang, F., Yao, G., Wu, J., Wu, H., Lian, Z., Geng, X., Chen, J., Yuan, Y., & Heng, P.-A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.14718 · Cited by 3
Seeing, Listening, Remembering, and Reasoning: A Multimodal Agent with Long-Term Memory
Long, L., He, Y., Ye, W., Pan, Y., Lin, Y., Li, H., Zhao, J., & Li, W. (2025) · arXiv · doi:10.48550/arxiv.2508.09736 · Cited by 102
“Most existing LVQA benchmarks [2, 9, 48, 60] mainly focus on visual understanding, such as action recognition and spatial/temporal perception, leaving a gap in evaluating higher-level cognitive abilities that rely on long-term memory and are crucial for real-world agents, such as understanding…”
Zhang, D., Zhao, Y., Huang, J., & Chen, Y. (2025) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2025.acl-long.222 · Cited by 2
SCVBench: A Benchmark with Multi-turn Dialogues for Story-Centric Video Understanding
You, S., Yuan, B., & Bao, B.-K. (2025) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2025/255 · Cited by 2
DToMA: Training-free Dynamic Token MAnipulation for Long Video Understanding
Yuan, B., You, S., & Bao, B. (2025) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2025/258 · Cited by 2
Activation Quantization of Vision Encoders Needs Prefixing Registers
Kim, S., Yeom, T., Kim, J., Park, W., Kim, K., & Lee, J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37520-9_2 · Cited by 2
Towards training-free long video understanding: methods, benchmarks, and open challenges
Liu, J., Wang, Y., Zhang, L., Wang, Y., Xu, S., Wang, L., Yan, J., Zhang, D., & Chen, X. (2025) · Vicinagearth. · doi:10.1007/s44336-025-00017-w · Cited by 2
FALCONEye: Finding Answers and Localizing Content in ONE-hour-long videos with multi-modal LLMs
Plou, C., Borja, C., Martínez-Cantín, R., & Murillo, A. C. (2026) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv61042.2026.00136 · Cited by 2
LVC: A Lightweight Compression Framework for Enhancing VLMS in Long Video Understanding
Wang, Z., Wu, H., Rong, Y. K., ZHAO, Y., Zhang, Y., & Xu, B. (2026) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp55912.2026.11462414 · Cited by 2
Sink-Token-Aware Pruning for Fine-Grained Video Understanding in Efficient Video LLMs
Kim, K., Kim, J., Min, K., Wang, Y., Moon, J., McAuley, J., & Park, C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37189-8_16 · Cited by 2
Incentivizing Vision Language Models to Search for Long Video Question Answering
Goel, H., Sharan, S. P., Shah, S., Choi, M.-K., An, J., Grauman, K. L., & Chinchali, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37450-9_29 · Cited by 2
Metadata
- Authors
- Zhou, J., Shu, Y., Zhao, B., Wu, B., Liang, Z., Xiao, S., Qin, M., Yang, X., Xiong, Y., Zhang, B., Huang, T., & Liu, Z.
- Affiliations
- Beijing University of Posts and Telecommunications (CN)
- Beijing Academy of Artificial Intelligence (CN)
- Shanghai Jiao Tong University (CN)
- Zhejiang University (CN)
- ORCID
- Venue
- 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
- Publisher
- IEEE
- Published
- 2025-06-10
- Pages
- 13691-13701
- ISSN
- 1063-6919 (ISSN-L)
- OpenAlex
- W4413144845
- Cited by
- 44
- References
- 0
- Citations by year
- 2026: 37 · 2025: 6
- FWCI
- 32.37
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by-nc-sa)
- arXiv (Cornell University)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Human Pose and Action Recognition
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Video Analysis and Summarization
- Keywords
- long video understanding, benchmark, multi-task evaluation, multimodal large language models, video length, video genres, surveillance video, egocentric video, animated cartoons, context length, image understanding
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
