Call for Papers
The papers submitted are non-archival, they will not be published in the CVPR Workshop Proceedings.
Important Dates
Submission Deadline: April 20, 2025
Review Deadline: April 30, 2025
Notification to authors: May 25, 2025
Camera Ready Submission: May 31, 2025
Topics of Interest
We invite submissions addressing various aspects of Video Large Language Models, including but not limited to the following areas:
Methods/Algorithms
Approaches for training MLLMs on video data, including designing training objectives, rewards, and improving efficiency in both training and inference.
Data Creation
Innovative strategies for leveraging web video data, advanced filtering techniques, synthetic data generation, long video understanding datasets and enriching datasets for video instruction tuning.
Evaluation and Analysis
Robust evaluation frameworks for existing models, focusing on improving interpretability, deriving novel insights, and introducing new metrics and benchmarks for VidLLMs.
Best Practices
Sharing best practices in training, evaluation and reproducibility, "bag-of-tricks" papers.
Applications
Creative applications of VidLLMs in addressing traditional Computer Vision (CV) tasks and beyond; Demonstrating practical implementations and real-world impact of MLLMs.
Comparison and Benchmarking
VidLLMs against expert computer vision models, highlighting strengths and areas for improvement.
Limitations, Risks and Safety
Addressing bias, fairness, and ethical challenges in VidLLMs, including factuality, hallucination, and safety concerns.
Emerging Research Areas
Longform video understanding with LLMs, efficient design for high-resolution video tasks, video grounding LLMs, and advancements in complex reasoning and agent designs for VidLLMs.
Submission Guidelines
We accept both short (4 pages) and long (8 pages) papers.
Apart from page count, submissions should follow the CVPR format. Please see the complete guidelines at: CVPR 2025 Author Instructions
Please submit the papers at: OpenReview
Program Committee
We have a diverse PC drawn from academia and industry. Each submission will receive at least two blind reviews.