MM-SpuBench: Towards Better Understanding of Spurious Biases in Multimodal LLMs

TL;DR

This paper introduces MM-SpuBench, a benchmark dataset designed to analyze and understand spurious biases in multimodal large language models, revealing their persistent reliance on such biases and challenges in mitigation.

Contribution

The work presents a new human-verified benchmark dataset for spurious biases in multimodal models and provides a comprehensive evaluation of current models' reliance on these biases.

Findings

Models persistently rely on spurious correlations.

Mitigation of biases remains challenging.

Benchmark reveals gaps in model robustness.

Abstract

Spurious bias, a tendency to exploit spurious correlations between superficial input attributes and prediction targets, has revealed a severe robustness pitfall in classical machine learning problems. Multimodal Large Language Models (MLLMs), which leverage pretrained vision and language models, have recently demonstrated strong capability in joint vision-language understanding. However, both the presence and severity of spurious biases in MLLMs remain poorly understood. In this work, we address this gap by analyzing the spurious biases in the multimodal setting and uncovering the specific inference-time data patterns that can manifest this problem. To support this analysis, we introduce MM-SpuBench, a comprehensive, human-verified benchmark dataset consisting of image-class pairs annotated with core and spurious attributes, grounded in our taxonomy of nine distinct types of spurious correlations. The benchmark is constructed using human-interpretable attribute information to capture a wide range of spurious patterns reflective of real-world knowledge. Leveraging this benchmark, we conduct a comprehensive evaluation of the state-of-the-art open-source and proprietary MLLMs with both standard accuracy and the proposed Conditional Generation Likelihood Advantage (CGLA). Our findings highlight the persistence of reliance on spurious correlations and the difficulty of mitigation on our benchmark. We hope this work can inspire new technical strides to mitigate these biases. Our benchmark is publicly available at https://huggingface.co/datasets/mmbench/MM-SpuBench.