Chengzhi Mao

Affiliations

Columbia University

papers

324

total citations

papers (12)

ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object

CVPR 2024arXiv

citations

Landscape Learning for Neural Network Inversion

ICCV 2023arXiv

citations

EditLord: Learning Code Transformation Rules for Code Editing

ICML 2025arXiv

citations

Multitask Learning Strengthens Adversarial Robustness

ECCV 2020arXiv

citations

What You Can Reconstruct From a Shadow

CVPR 2023

citations

Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision

NEURIPS 2025arXiv

citations

Chengzhi Mao

Affiliations

papers (12)

Generative Interventions for Causal Learning

Adversarial Attacks Are Reversible With Natural Supervision

SelfIE: Self-Interpretation of Large Language Model Embeddings

Causal Transportability for Visual Recognition

Doubly Right Object Recognition: A Why Prompt for Visual Rationales

Convolutional Visual Prompt for Robust Visual Perception

ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object

Landscape Learning for Neural Network Inversion

EditLord: Learning Code Transformation Rules for Code Editing

Multitask Learning Strengthens Adversarial Robustness

What You Can Reconstruct From a Shadow

Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision

papers (12)

Generative Interventions for Causal Learning

Adversarial Attacks Are Reversible With Natural Supervision

SelfIE: Self-Interpretation of Large Language Model Embeddings

Causal Transportability for Visual Recognition

Doubly Right Object Recognition: A Why Prompt for Visual Rationales

Convolutional Visual Prompt for Robust Visual Perception

ImageNet-D: Benchmarking Neural Network Robustness on Diffusion Synthetic Object

Landscape Learning for Neural Network Inversion

EditLord: Learning Code Transformation Rules for Code Editing

Multitask Learning Strengthens Adversarial Robustness

What You Can Reconstruct From a Shadow

Video Diffusion Models Excel at Tracking Similar-Looking Objects Without Supervision