Tsung-Yi Ho

papers

720

total citations

papers (12)

Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift

AAAI 2024arXiv

citations

Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations

CVPR 2023arXiv

citations

Uncovering and Quantifying Social Biases in Code Generation

NEURIPS 2023arXiv

citations

AutoVP: An Automated Visual Prompting Framework and Benchmark

ICLR 2024arXiv

citations

Be Your Own Neighborhood: Detecting Adversarial Examples by the Neighborhood Relations Built on Self-Supervised Learning

ICML 2024arXiv

citations

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

ICLR 2024arXiv

citations

CoP: Agentic Red-teaming for Large Language Models using Composition of Principles

NEURIPS 2025arXiv

citations

PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models

NEURIPS 2025

citations

Tsung-Yi Ho

papers (12)

RADAR: Robust AI-Text Detection via Adversarial Learning

MMA-Diffusion: MultiModal Attack on Diffusion Models

How to Backdoor Diffusion Models?

VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion Models

Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift

Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations

Uncovering and Quantifying Social Biases in Code Generation

AutoVP: An Automated Visual Prompting Framework and Benchmark

Be Your Own Neighborhood: Detecting Adversarial Examples by the Neighborhood Relations Built on Self-Supervised Learning

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

CoP: Agentic Red-teaming for Large Language Models using Composition of Principles

PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models

papers (12)

RADAR: Robust AI-Text Detection via Adversarial Learning

MMA-Diffusion: MultiModal Attack on Diffusion Models

How to Backdoor Diffusion Models?

VillanDiffusion: A Unified Backdoor Attack Framework for Diffusion Models

Elijah: Eliminating Backdoors Injected in Diffusion Models via Distribution Shift

Towards Compositional Adversarial Robustness: Generalizing Adversarial Training to Composite Semantic Perturbations

Uncovering and Quantifying Social Biases in Code Generation

AutoVP: An Automated Visual Prompting Framework and Benchmark

Be Your Own Neighborhood: Detecting Adversarial Examples by the Neighborhood Relations Built on Self-Supervised Learning

Rethinking Backdoor Attacks on Dataset Distillation: A Kernel Method Perspective

CoP: Agentic Red-teaming for Large Language Models using Composition of Principles

PermLLM: Learnable Channel Permutation for N:M Sparse Large Language Models