Probing Multimodal Large Language Models for Global and Local Semantic Representations

2024/02/27 by Mingxu Tao, Tao, Mingxu, Quzhe Huang +9 · 3 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.2402.17304

openalex publication_date 2024/02/27 · openalex created_date 2024/03/05 · openalex updated_date 2026/07/28

Abstract

The advancement of Multimodal Large Language Models (MLLMs) has greatly accelerated the development of applications in understanding integrated texts and images. Recent works leverage image-caption datasets to train MLLMs, achieving state-of-the-art performance on image-to-text tasks. However, there are few studies exploring which layers of MLLMs make the most effort to the global image information, which plays vital roles in multimodal comprehension and generation. In this study, we find that the intermediate layers of models can encode more global semantic information, whose representation vectors perform better on visual-language entailment tasks, rather than the topmost layers. We further probe models regarding local semantic representations through object recognition tasks. We find that the topmost layers may excessively focus on local information, leading to a diminished ability to encode global information. Our code and data are released via https://github.com/kobayashikanna01/probing_MLLM_rep.

Cited by

• RCI: A Score for Evaluating Global and Local Reasoning in Multimodal Benchmarks
• I-FailSense: Towards General Robotic Failure Detection with Vision-Language Models
• MIMIC: Multimodal Inversion for Model Interpretation and Conceptualization

Related

• Multimodal Large Language Models for Text-rich Image Understanding: A Comprehensive Review
• Cross-modal Information Flow in Multimodal Large Language Models
• Explaining Multi-modal Large Language Models by Analyzing their Vision Perception
• E5-V: Universal Embeddings with Multimodal Large Language Models
• CoF: Coarse to Fine-Grained Image Understanding for Multi-modal Large Language Models