Wangmeng Zuo

Full Professor@Harbin Institute of Technology·中国·OpenReview

研究方向

Open World Visual Understanding · Large Multimodal Model · Image and Video Generation · single image super-resolution · image deblurring · image denoising · Low level vision

Wangmeng Zuo

UniRestorer: Universal Image Restoration via Adaptively Estimating Image Degradation at Proper Granularity

PreferThinker: Reasoning-based Personalized Image Preference Assessment

Color3D: Controllable and Consistent 3D Colorization with Personalized Colorizer

MoSA: Motion-Coherent Human Video Generation via Structure-Appearance Decoupling

DeAltHDR: Learning HDR Video Reconstruction from Degraded Alternating Exposure Sequences

Better know nothing than half-know anything: A Precise and Efficient Dataset for Scientific Reasoning in Language Models

PhysWorld: From Real Videos to World Models of Deformable Objects via Physics-Aware Demonstration Synthesis

$1+1<1$? Breaking the Standalone Barrier in Federated Fine-Tuning of Multimodal Large Language Models under Non-IID Data

GLaVE-Cap: Global-Local Aligned Video Captioning with Vision Expert Integration

MIRAGE: Assessing Hallucination in Multimodal Reasoning Chains of MLLM

Visual-O1: Understanding Ambiguous Instructions via Multi-modal Multi-turn Chain-of-thoughts Reasoning

Federated Residual Low-Rank Adaptation of Large Language Models

RoomEditor: High-Fidelity Furniture Synthesis with Parameter-Sharing U-Net

Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuning

On the Importance of Language-driven Representation Learning for Heterogeneous Federated Learning

LLM as a Complementary Optimizer to Gradient Descent: A Case Study in Prompt Tuning

CPT: Consistent Proxy Tuning for Black-box Optimization