MM-CRITIC: A Holistic Evaluation of Large Multimodal Models as Multimodal Critique

TL;DR

This paper introduces MM-CRITIC, a comprehensive benchmark for evaluating the critique abilities of large multimodal models across various tasks and dimensions, using expert-informed scoring and extensive experiments.

Contribution

It presents the first holistic benchmark for multimodal critique evaluation, covering multiple task types and integrating expert-informed scoring for reliable assessment.

Findings

MM-CRITIC effectively evaluates critique capabilities of LMMs.

Response quality correlates with critique performance.

Critique difficulty varies across evaluation dimensions.

Abstract

The ability of critique is vital for models to self-improve and serve as reliable AI assistants. While extensively studied in language-only settings, multimodal critique of Large Multimodal Models (LMMs) remains underexplored despite their growing capabilities in tasks like captioning and visual reasoning. In this work, we introduce MM-CRITIC, a holistic benchmark for evaluating the critique ability of LMMs across multiple dimensions: basic, correction, and comparison. Covering 8 main task types and over 500 tasks, MM-CRITIC collects responses from various LMMs with different model sizes and is composed of 4471 samples. To enhance the evaluation reliability, we integrate expert-informed ground answers into scoring rubrics that guide GPT-4o in annotating responses and generating reference critiques, which serve as anchors for trustworthy judgments. Extensive experiments validate the effectiveness of MM-CRITIC and provide a comprehensive assessment of leading LMMs' critique capabilities under multiple dimensions. Further analysis reveals some key insights, including the correlation between response quality and critique, and varying critique difficulty across evaluation dimensions. Our code is available at https://github.com/MichealZeng0420/MM-Critic.