Yue Zhao

31

Papers

217

Total Citations

Papers (31)

Large Multilingual Models Pivot Zero-Shot Multimodal Learning across Languages

One-Minute Video Generation with Test-Time Training

Bayesian Diffusion Models for 3D Shape Reconstruction

Distilling Vision-Language Models on Millions of Videos

MetaOOD: Automatic Selection of OOD Detection Models

Edit Away and My Face Will not Stay: Personal Biometric Defense against Malicious Generative Editing

Distilling Structural Representations into Protein Sequence Models

Recognize Actions by Disentangling Components of Dynamics

FineGym: A Hierarchical Video Dataset for Fine-Grained Action Understanding

On Isometry Robustness of Deep 3D Point Cloud Models Under Adversarial Attacks

TSGCNet: Discriminative Geometric Feature Learning With Two-Stream Graph Convolutional Network for 3D Dental Model Segmentation

Revisiting Skeleton-Based Action Recognition

Learning Video Representations From Large Language Models

Temporal Action Detection With Structured Segment Networks

PointBA: Towards Backdoor Attacks in 3D Point Cloud

Omni-sourced Webly-supervised Learning for Video Recognition

BA-Net: Bridge Attention for Deep Convolutional Neural Networks

Real-Time Online Video Detection with Temporal Smoothing Transformers

Intra- and Inter-Action Understanding via Temporal Action Parsing

DPU: Dynamic Prototype Updating for Multimodal Out-of-Distribution Detection

Secure On-Device Video OOD Detection Without Backpropagation

RayZer: A Self-supervised Large View Synthesis Model

Preference Optimization for Molecule Synthesis with Conditional Residual Energy-based Models

VideoPrism: A Foundational Visual Encoder for Video Understanding

Position: TrustLLM: Trustworthiness in Large Language Models

Trajectory Convolution for Action Recognition

Automatic Unsupervised Outlier Model Selection

BOND: Benchmarking Unsupervised Outlier Node Detection on Static Attributed Graphs

ADBench: Anomaly Detection Benchmark

ADGym: Design Choices for Deep Anomaly Detection

Yinyang K-Means: A Drop-In Replacement of the Classic K-Means with Consistent Speedup