UPME: An Unsupervised Peer Review Framework for Multimodal Large Language Model Evaluation
This addresses the need for scalable and unbiased evaluation methods for MLLMs, reducing human workload in visual question answering tasks.
The paper tackles the problem of evaluating Multimodal Large Language Models (MLLMs) by proposing an unsupervised peer review framework that automatically generates questions and assesses answers using only image data, achieving Pearson correlations of 0.944 and 0.814 with human evaluations on two datasets.
Multimodal Large Language Models (MLLMs) have emerged to tackle the challenges of Visual Question Answering (VQA), sparking a new research focus on conducting objective evaluations of these models. Existing evaluation methods face limitations due to the significant human workload required to design Q&A pairs for visual images, which inherently restricts the scale and scope of evaluations. Although automated MLLM-as-judge approaches attempt to reduce the human workload through automatic evaluations, they often introduce biases. To address these problems, we propose an Unsupervised Peer review MLLM Evaluation framework. It utilizes only image data, allowing models to automatically generate questions and conduct peer review assessments of answers from other models, effectively alleviating the reliance on human workload. Additionally, we introduce the vision-language scoring system to mitigate the bias issues, which focuses on three aspects: (i) response correctness; (ii) visual understanding and reasoning; and (iii) image-text correlation. Experimental results demonstrate that UPME achieves a Pearson correlation of 0.944 with human evaluations on the MMstar dataset and 0.814 on the ScienceQA dataset, indicating that our framework closely aligns with human-designed benchmarks and inherent human preferences.