Wenguan Wang

Full Professor@Zhejiang University·中国·OpenReview

研究方向

AI agent · large language model · neural symbolic computing · semantic segmentation · embodied navigation · human-object interaction · video segmentation · human parsing · graph neural network · visual saliency

Wenguan Wang

Uncertainty-Aware 3D Reconstruction for Dynamic Underwater Scenes

Reconciling Visual Perception and Generation in Diffusion Models

Open-Set Semantic Gaussian Splatting SLAM with Expandable Representation

Hyperbolic Hierarchical Clustering for Visual Representation Learning

Uncertainty-Aware Gaussian Map for Vision-Language Navigation

Unbiased Object Detection Beyond Frequency with Visually Prompted Image Synthesis

SinkTrack: Attention Sink based Context Anchoring for Large Language Models

SEED-GRPO: Semantic Entropy Enhanced GRPO for Uncertainty-Aware Policy Optimization

Toward Spatial Intelligence: A Unified Self-supervised Framework for 3D Representation Learning from Unposed Multi-View Images

Moving Beyond Diffusion: Hierarchy-to-Hierarchy Autoregression for fMRI-to-Image Reconstruction

FairOCL: Fair Gradient Aggregation for Online Continual Learning

Composing Human Object Interaction with Decoupled Prototype for Zero-shot Learning

OmniGaze: Reward-inspired Generalizable Gaze Estimation in the Wild

Do as We Do, Not as You Think: the Conformity of Large Language Models

Learning Human-Object Interaction as Groups

Hydra-SGG: Hybrid Relation Assignment for One-stage Scene Graph Generation

Learning Clustering-based Prototypes for Compositional Zero-Shot Learning

Cycle-Consistent Learning for Joint Layout-to-Image Generation and Object Detection

Spatio-temporal Decoupled Knowledge Compensator for Few-Shot Action Recognition

HYBRID MODEL COLLABORATION FOR SIGN LANGUAGE TRANSLATION WITH VQ-VAE AND RAG ENHANCED LLMS

TAGA: Self-supervised Learning for Template-free Animatable Gaussian Avatars

Open-Ended 3D Metric-Semantic Representation Learning via Semantic-Embedded Gaussian Splatting

Towards Human-like Virtual Beings: Simulating Human Behavior in 3D Scenes

Generalization by Specialization: Unveiling Specialized Subnetworks in Large Language Models

MindGrapher: Dynamic-Aware fMRI-to-Video Reconstruction