Tao Huang

papers

1,848

total citations

papers (22)

Neighbor2Neighbor: Self-Supervised Denoising From Single Noisy Images

CVPR 2021arXiv

391

citations

Knowledge Distillation from A Stronger Teacher

NEURIPS 2022arXiv

361

citations

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

ICML 2025arXiv

214

citations

Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning

CVPR 2024arXiv

citations

GreedyNASv2: Greedier Search With a Greedy Path Filter

CVPR 2022arXiv

citations

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders

CVPR 2025arXiv

citations

Training-free Composite Scene Generation for Layout-to-Image Synthesis

ECCV 2024arXiv

citations

GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scene

ICCV 2025arXiv

citations

Learning Mask Invariant Mutual Information for Masked Image Modeling

ICLR 2025arXiv

citations

Parameterized Blur Kernel Prior Learning for Local Motion Deblurring

CVPR 2025

citations

A Unified Framework for Industrial Cel-Animation Colorization with Temporal-Structural Awareness

ICCV 2025

citations

SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures

ICCV 2025arXiv

citations

Low-Light Image Enhancement with Multi-Stage Residue Quantization and Brightness-Aware Attention

ICCV 2023

citations

Gain from Neighbors: Boosting Model Robustness in the Wild via Adversarial Perturbations Toward Neighboring Classes

CVPR 2025

citations

Tao Huang

papers (22)

Neighbor2Neighbor: Self-Supervised Denoising From Single Noisy Images

Knowledge Distillation from A Stronger Teacher

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba

Deep Gaussian Scale Mixture Prior for Spectral Compressive Imaging

GreedyNAS: Towards Fast One-Shot NAS With Greedy Supernet

Knowledge Diffusion for Distillation

Prioritized Architecture Sampling With Monto-Carlo Tree Search

FreeKD: Knowledge Distillation via Semantic Frequency Prompt

Diffusion Reward: Learning Rewards via Conditional Video Diffusion

DyRep: Bootstrapping Training With Dynamic Re-Parameterization

Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning

GreedyNASv2: Greedier Search With a Greedy Path Filter

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders

Training-free Composite Scene Generation for Layout-to-Image Synthesis

GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scene

Learning Mask Invariant Mutual Information for Masked Image Modeling

Parameterized Blur Kernel Prior Learning for Local Motion Deblurring

A Unified Framework for Industrial Cel-Animation Colorization with Temporal-Structural Awareness

SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures

Low-Light Image Enhancement with Multi-Stage Residue Quantization and Brightness-Aware Attention

Gain from Neighbors: Boosting Model Robustness in the Wild via Adversarial Perturbations Toward Neighboring Classes

papers (22)

Neighbor2Neighbor: Self-Supervised Denoising From Single Noisy Images

Knowledge Distillation from A Stronger Teacher

SparseVLM: Visual Token Sparsification for Efficient Vision-Language Model Inference

EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mamba

Deep Gaussian Scale Mixture Prior for Spectral Compressive Imaging

GreedyNAS: Towards Fast One-Shot NAS With Greedy Supernet

Knowledge Diffusion for Distillation

Prioritized Architecture Sampling With Monto-Carlo Tree Search

FreeKD: Knowledge Distillation via Semantic Frequency Prompt

Diffusion Reward: Learning Rewards via Conditional Video Diffusion

DyRep: Bootstrapping Training With Dynamic Re-Parameterization

Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning

GreedyNASv2: Greedier Search With a Greedy Path Filter

MoVE-KD: Knowledge Distillation for VLMs with Mixture of Visual Encoders

Training-free Composite Scene Generation for Layout-to-Image Synthesis

GLEAM: Learning Generalizable Exploration Policy for Active Mapping in Complex 3D Indoor Scene

Learning Mask Invariant Mutual Information for Masked Image Modeling

Parameterized Blur Kernel Prior Learning for Local Motion Deblurring

A Unified Framework for Industrial Cel-Animation Colorization with Temporal-Structural Awareness

SAM Encoder Breach by Adversarial Simplicial Complex Triggers Downstream Model Failures

Low-Light Image Enhancement with Multi-Stage Residue Quantization and Brightness-Aware Attention

Gain from Neighbors: Boosting Model Robustness in the Wild via Adversarial Perturbations Toward Neighboring Classes