vix.ing · top · new · best · stats

V*: Guided Visual Search as a Core Mechanism in Multimodal LLMs

2023/12/21 by Penghao Wu, Saining Xie, Wu, Penghao +1 · 2 voices · 215 citations
Computer Science · #Artificial intelligence #Benchmark (surveying) #Computer science #Focus (optics) #Geography #Human–computer interaction #Image Retrieval and Classification Techniques #Mechanism (biology) #Multimodal Machine Learning Applications #Process (computing) #Programming language #Semantic Web and Ontologies #Visual search

paper · pdf · doi:10.48550/arxiv.2312.14135

published in arXiv (Cornell University) (Cornell University)

openalex publication_date 2023/12/21 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

When we look around and perform complex tasks, how we see and selectively process what we see is crucial. However, the lack of this visual search mechanism in current multimodal LLMs (MLLMs) hinders their ability to focus on important visual details, especially when handling high-resolution and visually crowded images. To address this, we introduce V*, an LLM-guided visual search mechanism that employs the world knowledge in LLMs for efficient visual querying. When combined with an MLLM, this mechanism enhances collaborative reasoning, contextual understanding, and precise targeting of specific visual elements. This integration results in a new MLLM meta-architecture, named Show, sEArch, and TelL (SEAL). We further create V*Bench, a benchmark specifically designed to evaluate MLLMs in their ability to process high-resolution images and focus on visual details. Our study highlights the necessity of incorporating visual search capabilities into multimodal systems. The code is available https://github.com/penghao-wu/vstar.

Cited by

Discussions

Related