2025/06/05 by Lin, Jingyang, Jialian Wu, Wu, Jialian +18 · 9 citations
Computer Science · #Computation and Language (cs.CL) #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Human Pose and Action Recognition #Multimodal Machine Learning Applications
paper · pdf · doi:10.48550/arxiv.2506.05332
openalex publication_date 2025/06/05 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Recent long-form video-language understanding benchmarks have driven progress in video large multimodal models (Video-LMMs). However, the scarcity of well-annotated long videos has left the training of hour-long Video-LMMs underexplored. To close this gap, we present VideoMarathon, a large-scale hour-long video instruction-following dataset. This dataset includes around 9,700 hours of long videos sourced from diverse domains, ranging from 3 to 60 minutes per video. Specifically, it contains 3.3M high-quality QA pairs, spanning six fundamental topics: temporality, spatiality, object, action, scene, and event. Compared to existing video instruction datasets, VideoMarathon significantly extends training video durations up to 1 hour, and supports 22 diverse tasks requiring both short- and long-term video comprehension. Building on VideoMarathon, we propose Hour-LLaVA, a powerful and efficient Video-LMM for hour-scale video-language modeling. It enables hour-long video training and inference at 1-FPS sampling by leveraging a memory augmentation module, which adaptively integrates question-relevant and spatiotemporally informative semantics from the cached full video context. In our experiments, Hour-LLaVA achieves the best performance on multiple representative long video-language benchmarks, demonstrating the high quality of the VideoMarathon dataset and the superiority of the Hour-LLaVA model.