揭示多模态大模型中视觉与语言信息的流动路径。
Cross-modal Information Flow in Multimodal Large Language Models
- 分三层分析视觉与语言信息融合:底层传递整体图像特征,中层注入关键物体信息,高层生成最终预测。
- 在中间层,模型将与问题相关的目标视觉特征精准映射到对应词元位置。
- 适用于研究多模态模型内部机制、可解释性及信息编辑的科研人员。
近期自回归多模态大语言模型(MLLMs)在视觉-语言任务上展现出显著进展。尽管已有大量研究关注大语言模型中的语言处理,但对MLLMs内部工作机制以及视觉与语言信息如何交互仍知之甚少。本研究通过分析视觉问答任务中不同模态间的信息流,填补这一空白。以图像-问题对为输入,我们探究视觉与语言信息在模型中的融合位置与方式。基于LLaVA系列多个模型的实验发现,信息融合过程存在两个明显阶段:在低层,模型将整体图像的通用视觉特征传递至问题词元的表示;在中层,模型将与问题相关的特定物体视觉信息重新传递至问题中对应词元的位置;在高层,融合后的多模态表示被传播至输入序列末尾位置进行最终预测。结果揭示了多模态大模型在空间与功能上的处理结构,为未来多模态信息定位与编辑研究提供新视角。代码与数据集已公开:https://github.com/FightingFighting/cross-modal-information-flow-in-MLLM.git。
原文摘要 · Abstract (English)
The recent advancements in auto-regressive multimodal large language models (MLLMs) have demonstrated promising progress for vision-language tasks. While there exists a variety of studies investigating the processing of linguistic information within large language models, little is currently known about the inner working mechanism of MLLMs and how linguistic and visual information interact within these models. In this study, we aim to fill this gap by examining the information flow between different modalities -- language and vision -- in MLLMs, focusing on visual question answering. Specifically, given an image-question pair as input, we investigate where in the model and how the visual and linguistic information are combined to generate the final prediction. Conducting experiments with a series of models from the LLaVA series, we find that there are two distinct stages in the process of integration of the two modalities. In the lower layers, the model first transfers the more general visual features of the whole image into the representations of (linguistic) question tokens. In the middle layers, it once again transfers visual information about specific objects relevant to the question to the respective token positions of the question. Finally, in the higher layers, the resulting multimodal representation is propagated to the last position of the input sequence for the final prediction. Overall, our findings provide a new and comprehensive perspective on the spatial and functional aspects of image and language processing in the MLLMs, thereby facilitating future research into multimodal information localization and editing. Our code and collected dataset are released here: https://github.com/FightingFighting/cross-modal-information-flow-in-MLLM.git.
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。