发现多模态大模型依赖虚假视觉线索，导致识别错误和幻觉放大
SpurLens: Automatic Detection of Spurious Cues in Multimodal LLMs
- 用GPT-4与开放集检测器自动识别虚假视觉线索
- 移除虚假线索致准确率下降，幻觉强度提升超10倍
- 适合关注模型可靠性与鲁棒性研究者阅读
单模态视觉模型已知会依赖虚假相关性，但多模态大语言模型（MLLMs）在语言监督下是否同样存在此类偏差尚不明确。本文研究了MLLMs中的虚假偏差，提出SpurLens管道，利用GPT-4和开放集物体检测器在无监督条件下自动识别虚假视觉线索。研究发现，虚假相关性引发两大故障模式：(1) 过度依赖虚假线索进行物体识别，移除后准确率下降；(2) 物体幻觉被虚假线索放大逾10倍。我们在多种MLLMs和数据集上验证了该现象。此外，探索了提示集成与基于推理的提示等缓解策略，并通过消融实验分析虚假偏差的根本成因。本研究揭示了虚假相关性的持续存在，呼吁采用更严格的评估方法与缓解策略以提升MLLMs的可靠性。
原文摘要 · Abstract (English)
Unimodal vision models are known to rely on spurious correlations, but it remains unclear to what extent Multimodal Large Language Models (MLLMs) exhibit similar biases despite language supervision. In this paper, we investigate spurious bias in MLLMs and introduce SpurLens, a pipeline that leverages GPT-4 and open-set object detectors to automatically identify spurious visual cues without human supervision. Our findings reveal that spurious correlations cause two major failure modes in MLLMs: (1) over-reliance on spurious cues for object recognition, where removing these cues reduces accuracy, and (2) object hallucination, where spurious cues amplify the hallucination by over 10x. We validate our findings in various MLLMs and datasets. Beyond diagnosing these failures, we explore potential mitigation strategies, such as prompt ensembling and reasoning-based prompting, and conduct ablation studies to examine the root causes of spurious bias in MLLMs. By exposing the persistence of spurious correlations, our study calls for more rigorous evaluation methods and mitigation strategies to enhance the reliability of MLLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。