MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding

• Published in: arXiv.org 2024-07
• Main Authors: Wang, Fei, Fu, Xingyu, Huang, James Y, Li, Zekun, Liu, Qin, Liu, Xiaogeng, Ma, Mingyu Derek, Xu, Nan, Zhou, Wenxuan, Zhang, Kai, Yan, Tianyi Lorena, Mo, Wenjie Jacky, Hsiang-Hui, Liu, Pan, Lu, Li, Chunyuan, Xiao, Chaowei, Kai-Wei, Chang, Roth, Dan, Zhang, Sheng, Poon, Hoifung, Chen, Muhao
• Format: Article
• Language: English
• Subjects: Benchmarks; Hovering; Questions; Robustness; Scene analysis
• ISSN: 2331-8422

We introduce MuirBench, a comprehensive benchmark that focuses on robust multi-image understanding capabilities of multimodal LLMs. MuirBench consists of 12 diverse multi-image tasks (e.g., scene understanding, ordering) that involve 10 categories of multi-image relations (e.g., multiview, temporal relations). Comprising 11,264 images and 2,600 multiple-choice questions, MuirBench is created in a pairwise manner, where each standard instance is paired with an unanswerable variant that has minimal semantic differences, in order for a reliable assessment. Evaluated upon 20 recent multi-modal LLMs, our results reveal that even the best-performing models like GPT-4o and Gemini Pro find it challenging to solve MuirBench, achieving 68.0% and 49.3% in accuracy. Open-source multimodal LLMs trained on single images can hardly generalize to multi-image questions, hovering below 33.3% in accuracy. These results highlight the importance of MuirBench in encouraging the community to develop multimodal LLMs that can look beyond a single image, suggesting potential pathways for future improvements.