paper-with-me

홈 › Papers

MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding

2024-06-13 · Fei Wang, Xingyu Fu, James Y. Huang, Zekun Li, Qin Liu, Xiaogeng Liu, Mingyu Derek Ma, Nan Xu, Wenxuan Zhou, Kai Zhang, Tianyi Lorena Yan, Wenjie Jacky Mo, Hsiang-Hui Liu, Pan Lu, Chunyuan Li, Chaowei Xiao, Kai-Wei Chang, Dan Roth, Sheng Zhang, Hoifung Poon, Muhao Chen

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.

📄 PDF Abstract BibTeX arXiv:2406.09411

Code (1)

muirbench/MuirBench 공식 구현

Tasks

Multiple-choiceScene Understanding

Similar Papers 제목 키워드 기반

Decoding the Pulse of Reasoning VLMs in Multi-Image Understanding Tasks

2026-03-04 · Chenjun Li arxiv

Multi-image reasoning remains a significant challenge for vision-language models (VLMs). We investigate a previously overlooked phenomenon: during chain-of-thought (CoT) generation, the text-to-image (T2I) attention of r…

Enhancing Multi-Image Understanding through Delimiter Token Scaling

2026-02-02 · Minyoung Lee, Yeji Park, Dongjun Hwang, Yejin Kim 외 arxiv

Large Vision-Language Models (LVLMs) achieve strong performance on single-image tasks, but their performance declines when multiple images are provided as input. One major reason is the cross-image information leakage, w…

CGC: Compositional Grounded Contrast for Fine-Grained Multi-Image Understanding

2026-04-24 · Lihao Zheng, Zhenwei Shao, Yu Zhou, Yan Yang 외 arxiv

Although Multimodal Large Language Models (MLLMs) have advanced rapidly, they still face notable challenges in fine-grained multi-image understanding, often exhibiting spatial hallucination, attention leakage, and failur…

Image Attribution

Dynamic-VLM: Simple Dynamic Visual Token Compression for VideoLLM

2024-12-12 · Han Wang, Yuxiang Nie, YongJie Ye, Deng GuanYu 외

The application of Large Vision-Language Models (LVLMs) for analyzing images and videos is an exciting and rapidly evolving field. In recent years, we've seen significant growth in high-quality image-text datasets for fi…

Computational Efficiency

MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale

2024-12-06 · Jarvis Guo, Tuney Zheng, Yuelin Bai, Bo Li 외

Open-source multimodal large language models (MLLMs) have shown significant potential in a broad range of multimodal tasks. However, their reasoning capabilities remain constrained by existing instruction-tuning datasets…

Multimodal ReasoningVisual Question AnsweringVisual Question Answering (VQA)