paper-with-me

홈 › Papers

Cross-Modal Projection in Multimodal LLMs Doesn't Really Project Visual Attributes to Textual Space

2024-02-26 · Gaurav Verma, MinJe Choi, Kartik Sharma, Jamelle Watson-Daniels, Sejoon Oh, Srijan Kumar

Multimodal large language models (MLLMs) like LLaVA and GPT-4(V) enable general-purpose conversations about images with the language modality. As off-the-shelf MLLMs may have limited capabilities on images from domains like dermatology and agriculture, they must be fine-tuned to unlock domain-specific applications. The prevalent architecture of current open-source MLLMs comprises two major modules: an image-language (cross-modal) projection network and a large language model. It is desirable to understand the roles of these two modules in modeling domain-specific visual attributes to inform the design of future models and streamline the interpretability efforts on the current models. To this end, via experiments on 4 datasets and under 2 fine-tuning settings, we find that as the MLLM is fine-tuned, it indeed gains domain-specific visual capabilities, but the updates do not lead to the projection extracting relevant domain-specific visual attributes. Our results indicate that the domain-specific visual attributes are modeled by the LLM, even when only the projection is fine-tuned. Through this study, we offer a potential reinterpretation of the role of cross-modal projections in MLLM architectures. Project webpage: https://claws-lab.github.io/projection-in-MLLMs/

📄 PDF Abstract BibTeX arXiv:2402.16832

Code (1)

claws-lab/projection-in-mllms 공식 구현 pytorch

Tasks

Language ModelingLanguage ModellingLarge Language Model

Similar Papers 제목 키워드 기반

OneLLM: One Framework to Align All Modalities with Language

2023-12-06 · CVPR 2024 1 · Jiaming Han, Kaixiong Gong, Yiyuan Zhang, Jiaqi Wang 외

Multimodal large language models (MLLMs) have gained significant attention due to their strong multimodal understanding capability. However, existing works rely heavily on modality-specific encoders, which usually differ…

AllQuestion AnsweringVisual Question Answering

EmoLLM: Multimodal Emotional Understanding Meets Large Language Models

2024-06-24 · Qu Yang, Mang Ye, Bo Du

Multi-modal large language models (MLLMs) have achieved remarkable performance on objective multimodal perception tasks, but their ability to interpret subjective, emotionally nuanced multimodal content remains largely u…

Emotional Intelligence

AsymLoRA: Harmonizing Data Conflicts and Commonalities in MLLMs

2025-02-27 · Xuyang Wei, Chunlin Tian, Li Li

Effective instruction fine-tuning on diverse image-text datasets is crucial for developing a versatile Multimodal Large Language Model (MLLM), where dataset composition dictates the model's adaptability across multimodal…

Language ModelingLanguage ModellingLarge Language ModelMultimodal Large Language Model

ElasticMM: Efficient Multimodal LLMs Serving with Elastic Multimodal Parallelism

2025-07-14 · Zedong Liu, Shenggan Cheng, Guangming Tan, Yang You 외 arxiv

Multimodal large language models (MLLMs) extend LLMs to handle images, videos, and audio by incorporating feature extractors and projection modules. However, these additional components -- combined with complex inference…

LMFusion: Adapting Pretrained Language Models for Multimodal Generation

2024-12-19 · Weijia Shi, Xiaochuang Han, Chunting Zhou, Weixin Liang 외

We present LMFusion, a framework for empowering pretrained text-only large language models (LLMs) with multimodal generative capabilities, enabling them to understand and generate both text and images in arbitrary sequen…

Image Generationmultimodal generation