SPORTU: A Comprehensive Sports Understanding Benchmark for Multimodal Large Language Models

• Main Authors: Xia, Haotian, Yang, Zhengbang, Zou, Junbo, Tracy, Rhys, Wang, Yuqing, Lu, Chi, Lai, Christopher, He, Yanjun, Shao, Xun, Xie, Zhuoqing, Wang, Yuan-fang, Shen, Weining, Chen, Hanjie
• Format: Journal Article
• Language: English
• Published: 10.10.2024
• Subjects: Computer Science - Computation and Language; Computer Science - Computer Vision and Pattern Recognition

Multimodal Large Language Models (MLLMs) are advancing the ability to reason about complex sports scenarios by integrating textual and visual information. To comprehensively evaluate their capabilities, we introduce SPORTU, a benchmark designed to assess MLLMs across multi-level sports reasoning tas...