Abstract
Online information is increasingly consumed in video format. Much of this comes in the form of *raw video*: continuous footage taken on a cell phone, with a hand-held camera, or via CCTV, which is then directly uploaded to social media platforms and content sharing services. Whereas professional or even amateur-edited footage tends to feature scripted speech, chyrons, graphics, and metadata that help contextualize its subject matter, raw video typically contains none of these things, making it a much more challenging medium for information retrieval and machine understanding. To facilitate progress in this domain, we release MultiVENT-Raw, a multilingual collection of nearly 120,000 primarily raw videos (over 5,300 total hours), paired with 130 events and 222 event-centric queries, along with human-annotated video relevance judgments and human-extracted key facts for relevant videos. MultiVENT-Raw supports both a retrieval task---to identify videos in the collection relevant to a query event---and a generation task---to summarize event-related videos into a coherent report for a target user. We benchmark strong baselines on MultiVENT-Raw, showing both tasks to be challenging even for some of the latest multimodal models.
Keywords
Subject
Publication details
- Journal
- Not available
- Open access
- Green open access
Cite this article
APA 7
Kriz, R., Etter, D., Martin, A., Carpenter, C., Chakraborty, D., Recknor, H., Iranmanesh, R., Maciejewski, M., Murray, K., Yang, E., Van Durme, B., White, A. S., Yates, A., & Walden, W. (2026). MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos. https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos
MLA 9
Kriz, Reno, et al. "MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos." https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos.
Chicago (author–date)
Kriz, Reno, David Etter, Alexander Martin, Cameron Carpenter, Debashish Chakraborty, Hannah Recknor, Reihaneh Iranmanesh, Matthew Maciejewski, Kenton Murray, Eugene Yang, Benjamin Van Durme, Aaron Steven White, Andrew Yates, and William Walden. 2026. "MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos." https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos.
Harvard
Kriz, R., Etter, D., Martin, A., Carpenter, C., Chakraborty, D., Recknor, H., Iranmanesh, R., Maciejewski, M., Murray, K., Yang, E., Van Durme, B., White, A. S., Yates, A. and Walden, W. (2026) 'MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos', Available at: https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos.
Vancouver
Kriz R, Etter D, Martin A, Carpenter C, Chakraborty D, Recknor H, et al. MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos. https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos
IEEE
R. Kriz, D. Etter, A. Martin, C. Carpenter, D. Chakraborty, H. Recknor, R. Iranmanesh, M. Maciejewski, K. Murray, E. Yang, B. Van Durme, A. S. White, A. Yates, and W. Walden, "MultiVENT-Raw: A Benchmark for Retrieval and Reasoning over Raw Videos," https://omanscience.com/en/articles/multivent-raw-a-benchmark-for-retrieval-and-reasoning-over-raw-videos.