Zhihai He

Critical Sampling for Data-Driven Modeling of Unknown Dynamical Systems
Learning to Adapt CLIP for Few-Shot Monocular Depth Estimation
Cross-inferential networks for source-free unsupervised domain adaptation
BDC-Adapter Brownian Distance Covariance for Better Vision-Language Reasoning
Unsupervised Prototype Adapter for Vision-Language Models
Cycle optimization metric learning for few-shot classification
Critical Sampling for Robust Evolution Operator Learning of Unknown Dynamical Systems
Self-Correctable and Adaptable Inference for Generalizable Human Pose Estimation
Neuro-Modulated Hebbian Learning for Fully Test-Time Adaptation
Self-guided information for few-shot classification
Self-Constrained Inference Optimization on Structural Groups for Human Pose Estimation