Aug 27, 2025 · LMMS Progress Report: July 2025 Welcome back to our monthly series of LMMS Progress Reports! To those of you who participated in Best of LMMS this year, I hope you had …

Apr 21, 2025 · Inspired by this process, we propose DyFo (Dynamic Focus), a training-free dynamic focusing visual search method that enhances fine-grained visual understanding in …

This is the official repo for Dynamic Focus (Visual Search), a training-free visual search method for enhancing LMMs/MLLMs in Fine-Grained Visual Understanding by simulating human …

Geng Li, Jinglin Xu, Yunzhen Zhao and Yuxin Peng*, "DyFo: A Training-Free Dynamic Focus Visual Search for Enhancing LMMs in Fine-Grained Visual Understanding", 38th IEEE/CVF …

This repository contains the test code and evaluation data for DyFo: A Training-Free Dynamic Focus Visual Search for Enhancing LMMs in Fine-Grained Visual Understanding (CVPR 2025 …

Feb 20, 2024 · Moving beyond this, OpenAI's introduction of Large Multimodal Models (LMMs) represents a notable shift, enabling these models to process both images and textual data.

My research centers on multimodal foundation models and post-train them for solving long-horizon tasks. I also conduct research on multimodal understanding and generation.

Jun 19, 2025 · Recent advancements in large multimodal models (LMMs) provide a potential alternative because such models can respond to prompts that include both text and images …

I work toward long-context multimodal AI systems that can both understand and generate rich, interleaved streams of text, video, images, audio, code, and actions.

Apr 3, 2024 · Linear mixed models (LMMs) were used to map eQTLs across 38 cell types. In order to reliably estimate effect sizes for multiple cell types, multivariate adaptive shrinkage …