Question-Answering Dense Video Events
Hangyu Qin, Junbin Xiao, Angela Yao
Abstract
This paper presents question-answering on dense video events, a novel task that answers and grounds dense-event questions in long videos, thus challenging MLLMs to faithfully comprehend and reason about multiple events over extended periods of time. To facilitate the study, we construct DeVE-QA -- a dataset featuring 78K questions about 26K events on 10.6K long videos. Our benchmarking shows that state-of-the-art MLLMs struggle on DeVE-QA. For improvement, we propose DeVi, a novel training-free MLLM approach that highlights a hierarchical captioning module, a temporal event memory module, and a self-consistency checking module to respectively detect, contextualize and memorize, and ground dense-events in long videos for question answering. Extensive experiments show that DeVi is superior at answering dense-event questions and grounding relevant video moments. Compared with existing MLLMs, it achieves a notable increase of 4.8% and 2.1% for G(round)QA accuracy on DeVE-QA and NExT-GQA, respectively. Data and code are available at https://github.com/QHUni/DeVE-QA.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
4 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Towards One-to-Many Temporal Grounding cites Flamingo: a Visual Language Model for Few-Shot Learning
- InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning cites Flamingo: a Visual Language Model for Few-Shot Learning
- MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-Answering cites Flamingo: a Visual Language Model for Few-Shot Learning
- ViperGPT: Visual Inference via Python Execution for Reasoning cites Learning Transferable Visual Models From Natural Language Supervision
- ViperGPT: Visual Inference via Python Execution for Reasoning cites Flamingo: a Visual Language Model for Few-Shot Learning
- Flamingo: a Visual Language Model for Few-Shot Learning cites Learning Transferable Visual Models From Natural Language Supervision
- InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation cites Learning Transferable Visual Models From Natural Language Supervision
- InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and Generation cites Flamingo: a Visual Language Model for Few-Shot Learning
Cited by top-tier papers4
- MuKV: Multi-Grained KV Cache Compression for Long Streaming Video Question-AnsweringJunbin Xiao, Jiajun Chen, Tianxiang Sun, Xun Yang et al.CVPR 2026 · 6 citations
- Enrich and Detect: Video Temporal Grounding With Multimodal LlmsShraman Pramanick, Effrosyni Mavroudi, Yale Song, Rama Chellappa et al.ICCV 2025 · 4 citations
- Towards One-to-Many Temporal GroundingQi Xu, Tan Yue, Shihao Chen, Jiahao Meng et al.ICML 2026 · 1 citation
- Divid: Disentangled Spatial-Temporal Modeling within LLMs for Temporally Grounded Video UnderstandingYepeng Tang, Weining Wang, Longteng Guo, Tongtian Yue et al.ICLR 2026
Builds on31
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
- Flamingo: a Visual Language Model for Few-Shot LearningJean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech et al.NeurIPS 2022 · 6,707 citations
- InstructBLIP: Towards General-purpose Vision-Language Models with Instruction TuningWenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong et al.NeurIPS 2023 · 4,013 citations
- ViperGPT: Visual Inference via Python Execution for ReasoningDídac Surís, Sachit Menon, Carl VondrickICCV 2023 · 732 citations
- InternVid: A Large-scale Video-Text Dataset for Multimodal Understanding and GenerationYi Wang, Yinan He, Yizhuo Li, Kunchang Li et al.ICLR 2024 · 467 citations
Related papers
- Grounded Multi-Hop VideoQA in Long-Form Egocentric VideosQirui Chen, Shangzhe Di, Weidi XieAAAI 2025 · 35 citations
- A Simple LLM Framework for Long-Range Video Question-AnsweringCe Zhang, Taixi Lu, Md Mohaiminul Islam, Ziyang Wang et al.EMNLP 2024 · 37 citations
- MoReVQA: Exploring Modular Reasoning Models for Video Question AnsweringJuhong Min, Shyamal Buch, Arsha Nagrani, Minsu Cho et al.CVPR 2024 · 27 citations
- MIST : Multi-modal Iterative Spatial-Temporal Transformer for Long-form Video Question AnsweringDifei Gao, Luowei Zhou, Lei Ji, Linchao Zhu et al.CVPR 2023
- Grounded-VideoLLM: Sharpening Fine-grained Temporal Grounding in Video Large Language ModelsHaibo Wang, Zhiyang Xu, Yu Cheng, Shizhe Diao et al.EMNLP Findings 2025 · 14 citations
