用反事实方法揭示大型视觉语言模型中的偏见
Uncovering Bias in Large Vision-Language Models with Counterfactuals
AI总结：
本文通过反事实图像变化研究大型视觉语言模型中的社会偏见，发现输入图像中的种族、性别等属性会显著影响生成文本的毒性和能力相关词汇。
摘要：
随着大型语言模型（LLMs）具备日益强大的能力，许多大型视觉语言模型（LVLMs）被提出以将视觉输入引入LLMs。此类模型在生成文本时同时以输入图像和文本提示为条件，支持视觉问答和多模态聊天等多种应用场景。虽然已有研究考察了LLMs生成文本中包含的社会偏见，但这一主题在LVLMs中相对未被探索。由于文本和视觉模态所包含的信息会混淆地共同诱发偏见，考察LVLMs中的社会偏见尤其具有挑战性。为解决这一难题，我们对不同LVLMs在输入图像发生反事实变化时生成的文本进行了大规模研究。具体而言，我们向LVLMs提供相同的开放式文本提示，同时以来自不同反事实集合的图像为条件，其中每个集合包含在描绘共同主体（如医生）方面基本相同的图像，仅在交叉社会属性（如种族和性别）上有所不同。我们在这种反事实生成设置下全面评估了不同LVLMs生成的文本，发现输入图像中描绘的种族、性别和身体特征等社会属性会显著影响毒性以及能力相关词汇的生成。
英文摘要：
With the advent of Large Language Models (LLMs) possessing increasingly impressive capabilities, a number of Large Vision-Language Models (LVLMs) have been proposed to augment LLMs with visual inputs. Such models condition generated text on both an input image and a text prompt, enabling a variety of use cases such as visual question answering and multimodal chat. While prior studies have examined the social biases contained in text generated by LLMs, this topic has been relatively unexplored in LVLMs. Examining social biases in LVLMs is particularly challenging due to the confounding contributions of bias induced by information contained across the text and visual modalities. To address this challenging problem, we conduct a large-scale study of text generated by different LVLMs under counterfactual changes to input images. Specifically, we present LVLMs with identical open-ended text prompts while conditioning on images from different counterfactual sets, where each set contains images which are largely identical in their depiction of a common subject (e.g., a doctor), but vary only in terms of intersectional social attributes (e.g., race and gender). We comprehensively evaluate the text produced by different LVLMs under this counterfactual generation setting and find that social attributes such as race, gender, and physical characteristics depicted in input images can significantly influence toxicity and the generation of competency-associated words.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
