MOSSBench: Is Your Multimodal Language Model Oversensitive to Safe Queries?

TL;DR

This paper reveals that advanced multimodal language models often reject harmless queries due to oversensitivity triggered by specific visual stimuli, highlighting a need for improved safety mechanisms.

Contribution

It introduces MOSSBench, a benchmark with 300 benign multimodal queries, to systematically evaluate oversensitivity in 20 state-of-the-art MLLMs, revealing prevalent overcaution issues.

Findings

Oversensitivity is common, with refusal rates up to 76%.

Safer models tend to be more oversensitive.

Different stimuli cause errors at perception, reasoning, and safety judgment stages.

Abstract

Humans are prone to cognitive distortions -- biased thinking patterns that lead to exaggerated responses to specific stimuli, albeit in very different contexts. This paper demonstrates that advanced Multimodal Large Language Models (MLLMs) exhibit similar tendencies. While these models are designed to respond queries under safety mechanism, they sometimes reject harmless queries in the presence of certain visual stimuli, disregarding the benign nature of their contexts. As the initial step in investigating this behavior, we identify three types of stimuli that trigger the oversensitivity of existing MLLMs: Exaggerated Risk, Negated Harm, and Counterintuitive Interpretation. To systematically evaluate MLLMs' oversensitivity to these stimuli, we propose the Multimodal OverSenSitivity Benchmark (MOSSBench). This toolkit consists of 300 manually collected benign multimodal queries, cross-verified by third-party reviewers (AMT). Empirical studies using MOSSBench on 20 MLLMs reveal several insights: (1). Oversensitivity is prevalent among SOTA MLLMs, with refusal rates reaching up to 76% for harmless queries. (2). Safer models are more oversensitive: increasing safety may inadvertently raise caution and conservatism in the model's responses. (3). Different types of stimuli tend to cause errors at specific stages -- perception, intent reasoning, and safety judgement -- in the response process of MLLMs. These findings highlight the need for refined safety mechanisms that balance caution with contextually appropriate responses, improving the reliability of MLLMs in real-world applications. We make our project available at https://turningpoint-ai.github.io/MOSSBench/.