The Workshop on Video Large Language Models (VidLLMs) sponsored by Amazon Science has been accepted to CVPR 2025 that will take place in Nashville, TN.

The VidLLMs Workshop focuses on the latest advancements and challenges of Video Large Language Models. VidLLMs find applications across various fields where video content plays a crucial role. In educational technology, they can create interactive learning experiences by responding to educational videos with explanations or summaries. In healthcare, VidLLMs could assist in training or simulations by providing real-time insights and feedback. They could also play a crucial role in flexible sports analytics systems. Additionally, VidLLMs can enhance user interaction in customer service and entertainment through dynamic, video-based interfaces. This development of VidLLMs raises various open research questions and directions that this workshop seeks to provide a platform for discussing and analyzing.

Topics of Interest

• Methods and algorithms for training Video-LLMs
• Data creation and curation
• Evaluation and analysis (metrics, benchmarks)
• Applications in education, healthcare, sports analytics, etc.
• Comparisons with expert computer vision models
• Limitations, risks, and safety considerations
• Emerging research areas (e.g. multilingual, compositional reasoning)

Please visit the workshop home page for full details, https://www.crcv.ucf.edu/cvpr2025-vidllms-workshop/