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 301–400 of 454
Weaver: End-to-End Agentic System Training for Video Interleaved Reasoning
Shi, Y., Di, S., Chen, Q., Wang, Q., Cai, J., Jiang, X., Hu, Y., & Xie, W. (2026) · doi:10.1109/cvprf71909.2026.00887 · Cited by 2
Design Choices for Extending the Context Length of Visual Language Models
Li, M., Li, L., Gong, S., & Liu, Q. (2025) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2025.acl-long.1603 · Cited by 2
EMCompress: Video-LLMs with Endomorphic Multimodal Compression
Fan, Z., Liu, J., Zhang, Y., Wang, Z., Fung, Y. R., Li, M., & Ji, H. (2026) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2026.findings-acl.8 · Cited by 2
Li, Y., Liu, Z., Kong, Y., Li, G., Zhang, J., Bian, C., Liu, F., Yao, L., & Sun, Z. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2501.16786 · Cited by 2
Self-citationSparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMsPreprint
Zhang, X., Du, C., Yu, S., Wu, J., Zhang, F., Gao, W., & Liu, Q. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.19155 · Cited by 2
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/v1 · Cited by 2
CacheFlow: Compressive Streaming Memory for Efficient Long-Form Video UnderstandingPreprint
Patel, S., & Patel, D. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.13644 · Cited by 2
JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in RoboticsPreprint
Jahangard, S., Mohammadi, M., Shen, Y., Cai, Z., & Rezatofighi, H. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.10287 · Cited by 2
Enabling Disaggregated Multi-Stage MLLM Inference via GPU-Internal Scheduling and Resource SharingPreprint
Zhao, L., Zhou, H., Che, Y., & Cheng, D. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.17574 · Cited by 2
VideoScaffold: Elastic-Scale Visual Hierarchies for Streaming Video Understanding in MLLMsPreprint
Zheng, N., Huang, J., Guo, Q., & Zhao, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.22226 · Cited by 2
TiFRe: Text-guided Video Frame Reduction for Efficient Video Multi-modal Large Language ModelsPreprint
Zheng, X., Wang, Z., & Peng, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.08861 · Cited by 2
RT-NeuS: Towards Real-Time Neuro-Symbolic Video Understanding via Adaptive Temporal VerificationPreprint
Liang, S., Shah, S., Zhou, C., Sharan, S. P., Goel, H., Sanyal, A., Chinchali, S., & Datta, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2602.23553 · Cited by 2
WAT: Online Video Understanding Needs Watching Before ThinkingPreprint
Han, Z., Sun, H., Xu, J., Tang, C., Lei, Y., Zhang, X., Sun, H., He, Z., & Sun, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.13412 · Cited by 2
One Token per Highly Selective Frame: Towards Extreme Compression for Long Video UnderstandingPreprint
Zhang, Z., Pang, Z., Chen, S., Hao, X., Bhat, V., & Wang, Y.-X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.14149 · Cited by 2
Valley3: Scaling Omni Foundation Models for E-commercePreprint
Chen, Z., Zhou, G., Yang, M., Yin, Q., Zhao, Z., Yao, H., Xia, P., Chen, C., & Qiu, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.01278 · Cited by 2
Zhang, P., Zhang, G., He, W., Zhang, L., Liu, M., Xia, Y., Peng, Z., Dai, W., Liu, J., Tang, H., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.26680 · Cited by 2
Sandboxed Coding Agents are Competitive Omni-modal Task SolversPreprint
Chen, D., Huang, X., Hu, Z., Shi, Q., Li, D., & Zhou, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.00579 · Cited by 2
TimeThink: Reasoning with Time for Video LLMsPreprint
Li, H., Guo, L., Liu, Z., Hao, D., Tang, Y., Zhao, Z., Jiang, J., Jin, Z., Chen, C., Ҫirkinoğlu, H. O., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.05089 · Cited by 2
Self-citationJung, W., & Kim, J. (2025) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/v1/2025.findings-emnlp.1340 · Cited by 1
MovieCORE: COgnitive REasoning in Movies
Faure, G. J., Chen, M.-H., Yeh, J.-F., Cheng, Y., Su, H.-T., Tang, Y.-H., Lai, S., & Hsu, W. H. (2025) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2025.emnlp-main.66 · Cited by 1
CASA: Cross-Attention over Self-Attention for Efficient Vision-Language Fusion
Böhle, M., Royer, A., Marrie, J., Grave, É., & Pérez, P. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37235-2_31 · Cited by 1
Attend and Replay: Efficient Action Understanding in Long Videos via Mechanistic Interpretability
Hou, P., Zhang, J., & Huang, D. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00688 · Cited by 1
EXPLORE-Bench: Egocentric Scene Prediction with Long-Horizon Reasoning
Yu, C., Zhu, X., Du, C., Yu, P., Zhai, W., Cao, Y., & Zha, Z.-J. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37335-9_36 · Cited by 1
QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension
Luo, Y., Chen, W., Huang, W., Yin, S., Lin, H., Huang, J., Fu, C., Ji, J., Zheng, X., & Luo, J. (2026) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v40i29.39595 · Cited by 1
JRDB-Reasoning: A Difficulty-Graded Benchmark for Visual Reasoning in Robotics
Jahangard, S., Mohammadi, M., Shen, Y., Cai, Z., & Rezatofighi, H. (2026) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v40i7.37443 · Cited by 1
Towards Temporal Compositional Reasoning in Long-Form Sports Videos
Cao, S., Zhang, L., Zeng, R., & Liu, Z.-Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37419-6_4 · Cited by 1
Self-citationNEST: Narrative Event Structures in Time for Long Video Understanding
Asgarov, A., Narasimhan, K., Sarker, N. H., Alomari, H., Tang, C.-W., Sivakumar, A., Hakim, Z. I. A., Mallampati, S., & Thomas, C. (2026) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/2026.emnlp-main.2512 · Cited by 1
MVP: Enhancing Video Large Language Models via Self-supervised Masked Video Prediction
Sun, X., Wu, Z., Ding, Z., & Xu, L. (2026) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2026.acl-long.891 · Cited by 1
ReQuest: Rethinking-Based Question-Aware Frame Selection for Long-Form Video QA
Kim, M., Yun, S., Kim, Y. T., Moon, J., Choi, J. H., & Kim, S. T. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37235-2_23 · Cited by 1
LENS: Adaptive Spatio-Temporal Zooming for Keyframe Sampling in Long-Form Videos
Zhang, C., He, J., Sycara, K. P., & Xie, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37095-2_28 · Cited by 1
CrossView: Can Vision-Language Models Reason Across Cameras?
Shah, S., Sharan, S. P., Goel, H., Pasula, M., Hebbalae, A., Choi, M., & Chinchali, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37464-6_15 · Cited by 1
HAP: Head-Adaptive Visual Token Pruning via Cross-Modal Alignment
Sun, Y., Ji, H., Jin, Y., Deng, C., Fu, L., & Wang, X. (2026) · Findings of the Association for Computational Linguistics: EMNLP (Findings of EMNLP) · doi:10.18653/2026.findings-emnlp.1450 · Cited by 1
An Evidence Framework for Event-Grounded Multimodal Retrieval and Video Question Answering
Yuan, H., Lin, P., Wang, Z., & Li, R. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3838690 · Cited by 1
VideoSAVi: Self-Aligned Video Language Models without Human Supervision
Kulkarni, Y., & Fazli, P. (2024) · PubMed Central · doi:10.48550/arxiv.2412.00624 · Cited by 1
Espresso: High Compression For Rich Extraction From Videos for Your Vision-Language ModelPreprint
Yu, K. P., Dave, A., Ambruş, R., & Mercat, J. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.04729 · Cited by 1
IQViC: In-context, Question Adaptive Vision Compressor for Long-term Video Understanding LMMsPreprint
Yamao, S., Miyahara, N., Harazono, Y., & Takeuchi, S. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2412.09907 · Cited by 1
iLLaVA: An Image is Worth Fewer Than 1/3 Input Tokensin Large Multimodal ModelsPreprint
Hu, L., Shang, F., Wan, L., & Feng, W. (2024) · Qeios · doi:10.32388/69kflk · Cited by 1
Clapper: Compact Learning and Video Representation in VLMsPreprint
Kong, L., Zhang, H., Zhang, J., Huang, J., Li, K., Wang, Q., & Zhang, F. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2505.15529 · Cited by 1
Evaluating point-light biological motion in multimodal large language modelsPreprint
Kadambi, A., Iacoboni, M., Aziz‐Zadeh, L., & Narayanan, S. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.23517 · Cited by 1
LeanPO: Lean Preference Optimization for Likelihood Alignment in Video-LLMsPreprint
Wang, X., Huang, J., Li, Y., & Peng, P. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.05260 · Cited by 1
PAVE: Patching and Adapting Video Large Language ModelsPreprint
Liu, Z., Li, Y., Nguyen, K. D., Zhong, Y., & Li, Y. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2503.19794 · Cited by 1
Self-citationVITAL: Vision-Encoder-centered Pre-training for LMMs in Visual Quality AssessmentPreprint
Jia, Z., Cao, L., Han, J., Zhang, Z., Qian, J., Wang, J., Chen, Z., Zhai, G., & Min, X. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2511.17962 · Cited by 1
Mosaic: Cross-Modal Clustering for Efficient Video UnderstandingPreprint
Wang, T., Zhou, H., Song, C., Li, Q., & Ren, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.10060 · Cited by 1
Progressive Online Video Understanding with Evidence-Aligned Timing and Transparent DecisionsPreprint
Zhang, K., Yang, Z., Han, M., Hao, H., Zhuge, Y., Li, C., Zhao, J., Li, Z., & Chang, X. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.18459 · Cited by 1
Ma, M. Q., Guo, W., Agrawal, A., Gupta, A., Liang, P. P., Salakhutdinov, R., & Morency, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.01662 · Cited by 1
CoRDS: Coreset-based Representative and Diverse Selection for Streaming Video UnderstandingPreprint
Mahdizadeh, A., Moghadam, P. A., Li, M., He, X., & Sigal, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.14310 · Cited by 1
Efficient Long-Context Modeling in Diffusion Language Models via Block Approximate Sparse AttentionPreprint
Zhang, W., Wu, Y., Wang, H., Liu, Y., Dou, H., Yang, S., Wu, S., Zhao, H., & Jia, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.19726 · Cited by 1
TeachObs: A Human-Validated Benchmark for Multimodal Teaching Observation and Model EvaluationPreprint
Jeong, Y., Yoo, Y., Bae, J., Sohn, S., Han, H., Lee, J., Scott, H., & Lee, U. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.30673 · Cited by 1
Wang, H., & Huang, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.05833 · Cited by 1
TuringViT: Making SOTA Vision Transformers Accessible to AllPreprint
Wu, Q., Chen, H., Chen, L., Xin, R., Zheng, J., Wang, M., Hu, J., Zhu, D., Ma, Y., Wei, Y., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.24253 · Cited by 1
From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQAPreprint
Korkut, S., Sarmiento, M. A. B., Kim, S., & Akata, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.30220 · Cited by 1
刘云峰, Yang, Y., Han, J., Huang, Z., Tang, Y., Zeng, X., Wu, G., & Wang, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.14660 · Cited by 1
严彩君, Zhou, Y., Shi, M., Sun, H., Li, Y., Cai, Y., & Jiang, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.08612 · Cited by 1
TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live StreamingPreprint
Hu, Y., Qian, Y., Gu, M., Tao, Y., Chen, Y., Luo, Y., Liu, Z., Jin, M., & Ma, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.20958 · Cited by 1
Self-citationCoarse Indexing, Fine Evidence: Decoupling Temporal Granularity in Long-Video RAGPreprint
Jin, Z., Lin, Z., Zheng, B., Fang, J., & Yang, H. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.23011 · Cited by 1
Guo, Y., Wang, T., Zhang, Q., Meng, L., Zhang, Y., Cao, J., Jiang, H., Wu, C., Wu, J., Chen, S., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.05916 · Cited by 1
TRACE: Temporal Audit and Condition-aware Evaluation of Streaming Video UnderstandingPreprint
Ma, Y., Zhang, Q., Liu, P., & Zhao, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.30670 · Cited by 1
CASE: Cost-Aware Stopping for Efficient Long-Video AgentsPreprint
Du, Y., Liu, C., Liu, Z., Zheng, F., Wang, Z., Nie, J., & Huang, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.05400 · Cited by 1
Self-citationYuan, L., Wang, J., Sun, H., Zhang, Y., & Lin, Y. (2025) · arXiv · doi:10.48550/arxiv.2501.07888
Shifting AI Efficiency From Model-Centric to Data-Centric Compression
Liu, X., Wen, Z., Wang, S., Chen, J., Tao, Z., Wang, Y., Chen, T., Jin, X., Zou, C., Wang, Y., et al. (2025) · arXiv · doi:10.48550/arxiv.2505.19147
Sparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMs
Zhang, X., Du, C., Yu, S., Wu, J., Zhang, F., Gao, W., & Liu, Q. (2025) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2025.acl-short.59
DisCo: Towards Distinct and Coherent Visual Encapsulation in Video MLLMs
Zhao, J., Zheng, R., Wang, Y., Wang, H., & Zhao, H. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02016
All-in-one: Understanding and Generation in Multimodal Reasoning with the MAIA BenchmarkPreprint
Testa, D., Bonetta, G., Bernardi, R., Bondielli, A., Lenci, A., Miaschi, A., Passaro, L., & Magnini, B. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2502.16989
SD-MVSum: Script-Driven Multimodal Video Summarization Method and Datasets
Mylonas, M., Zerva, C., Apostolidis, E., & Mezaris, V. (2026) · IEEE Access · doi:10.1109/access.2026.3746362
EVRAG: Enhanced Video Retrieval-Augmented Generation for Democratized Social Computing
Yu, J., Farag, M., & Li, Y. (2025) · Lecture notes in computer science · doi:10.1007/978-3-032-07715-8_14
Kim, G., & Seo, M. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.13564
Multimodal Long Video Modeling Based on Temporal Dynamic ContextPreprint
Hao, H., Han, J., Zhang, Y., & Yue, X. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2504.10443
Task-Aware KV Compression For Cost-Effective Long Video UnderstandingPreprint
Qin, M., Shu, Y., Zhang, P., Lun, K., Yuan, H., Zhou, J., Xiao, S., Zhao, B., & Liu, Z. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2506.21184
Self-citationReasoning Like Experts: Leveraging Multimodal Large Language Models for Drawing-based Psychoanalysis
Ma, X., Jiang, Y., Erfani, S., Bailey, J., Liu, W., Ehinger, K. A., & Lau, J. H. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3755744
Accelerating Long Video Understanding via Compressed Scene Graph-Enabled Chain-of-Thought
Ling, T., Shi, S., & Wang, D. (2025) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3746027.3754765
ViPE: Visual Perception in Parameter Space for Efficient Video-Language Understanding
Lu, S., Yue, T., Guo, L., Li, H., He, X., Liu, S., & Liu, J. (2025) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/v1/2025.emnlp-main.897
Wang, K., Zheng, Q., Xie, J., Zhang, Y., Luo, J., Lu, H., Lin, L., Zhou, F., & Li, G. (2025) · IEEE/CVF International Conference on Computer Vision (ICCV) · doi:10.1109/iccv51701.2025.02089
MovieCORE: COgnitive REasoning in MoviesPreprint
Faure, G. J., Chen, M.-H., Yeh, J.-F., Cheng, Y., Su, H.-T., Tang, Y.-H., Lai, S.-H., & Hsu, W. H. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2508.19026
MESH -- Understanding Videos Like Human: Measuring Hallucinations in Large Video ModelsPreprint
Yang, G., Chen, Z., Wong, M. H., Lei, H., Chen, Y., Li, Z., Zhou, K., & Cheng, J. S.-C. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2509.08538
Cao, Z., He, Y., Liu, A., Chen, F., Wang, Z., & Xie, J. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2512.23483
LEMON: How Well Do MLLMs Perform Temporal Multimodal Understanding on Instructional Videos?Preprint
Yu, Z., Shen, L., Zhao, J., & Sun, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2601.20705
Singh, M., Challagundla, J., Raina, S., & Jarsania, J. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-6950-2_40
A Taxonomy of Knowledge Bases for Retrieval-Augmented Methods in Vision: A Comprehensive Survey
Kim, G., Lee, D., & Yoo, J. (2026) · IEEE Access · doi:10.1109/access.2026.3668187
Less Data, Faster Convergence: Goal-Driven Data Optimization for Multimodal Instruction TuningPreprint
Wu, R., Zhao, H., Ci, H., & Wang, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.12478
A Multi-Agent Perception-Action Alliance for Efficient Long Video ReasoningPreprint
Xu, Y., Liu, G., Kompella, R. R., Huang, T., Hu, S., İlhan, F., Tekin, S. F., Yahn, Z., & Liu, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.14052
Self-citationKim, G., & Seo, M. (2026) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v40i7.37485
ForeSea: AI Forensic Search with Multi-modal Queries for Video Surveillance
Park, H., Li, Y., Cho, J., Choi, S., Lee, J., Jing, T., Zhang, S., Hayat, M., Gao, D., Bi, N., et al. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37461-5_10
Rewind-Augmented Generation for Long Video Understanding
Huang, J., Chen, L., Liang, T., Zeng, X., Kong, M. C., Li, Y., & Zhu, Q. (2026) · IEEE Transactions on Multimedia · doi:10.1109/tmm.2026.3679059
Gao, L., Ma, Y., Bi, S., & Li, S. (2026) · Scientific Reports · doi:10.1038/s41598-026-43987-7
AdaSpark: Adaptive Sparsity for Efficient Long-Video UnderstandingPreprint
Li, H., Liu, Z., Guo, L., Yue, T., Tang, Y., Zhu, X., Zheng, C., Wang, Z., Wang, Z., Song, J., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.08077
Self-citationEnhancing MLLMs for Online Understanding in Video Services via Preference Optimization
Qi, Q., He, Y., Zhang, M., Sun, H., Ren, P., Wang, H., Liao, J., & Wang, J. (2026) · IEEE Transactions on Services Computing · doi:10.1109/tsc.2026.3687026
StoryTR: Narrative-Centric Video Temporal Retrieval with Theory of Mind ReasoningPreprint
Zhong, X., Xie, Y., Bi, G., Yang, J., & Chen, G. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.23198
SYNCR: A Cross-Video Reasoning Benchmark with Synthetic GroundingPreprint
Ghazanfari, S., Garg, S., Krishnamurthy, P., & Khorrami, F. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.08412
Dai, Y., Jiao, D., & Lin, T. (2026) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3767308.3838642
SVI-Bench: A Dynamic Microworld for Strategic Video Intelligence
Pan, Y., Han, Y., Ha, S., Islam, M. M., Zhang, B., Torresani, L., & Bertasius, G. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37095-2_12
Self-citationLinear Scaling Video VLMs for Long Video Understanding
Eyzaguirre, C., Wu, J., & Niebles, J. C. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37369-4_13
Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVRPreprint
Qin, C., Yang, C., Si, Q., Gu, N., Fu, P., & Lin, Z. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2606.03087
Chen, Z., Wang, W., Zhu, J., Wang, X., Liu, Y., Liu, Z. T., Luo, G., Sun, P., Su, W., Tian, H., et al. (2026) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2026.3702168
Self-citationReflective Cross-Granularity Grounding with Preference Optimization for Long Video Understanding
Feng, W., Wang, X. Y., Chen, H., Zhan, Y.-W., Song, Z., Huang, B., Zheng, K., & Zhu, W. (2026) · International Conference on Multimedia Retrieval · doi:10.1145/3805622.3810717
Rauf, A., Hasssan, A., & Slabaugh, G. (2026) · Conference on Empirical Methods in Natural Language Processing (EMNLP) · doi:10.18653/2026.emnlp-main.2377
Wang, D., Shi, J., Lu, Y., Chen, J., & Du, C. (2026) · Drones · doi:10.3390/drones10070491
VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context
Shen, X., & Elhoseiny, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37038-9_36
Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction
Fan, S., Liu, Q., Yin, R., Guo, M.-H., & Yang, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37362-5_14
MoPrune: Scene-Guided Motion-Aware Token Pruning for Efficient Video Large Language Models
Hong, W., Wang, Z., Zhang, Y., & Wang, Z. (2026) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2026.findings-acl.344
LongVQUBench: Benchmarking Long-Term Video Quality Understanding of Vision-Language Models
Nema, A., Zhu, H., Zhang, X., & Lin, W. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37138-6_10
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
