Chao Du

Affiliations

Alibaba GroupSea AI LabTsinghua University

papers

1,951

total citations

papers (22)

Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast

ICML 2024arXiv

103

citations

When Attention Sink Emerges in Language Models: An Empirical View

ICLR 2025arXiv

citations

Weak-to-Strong Jailbreaking on Large Language Models

ICML 2025arXiv

citations

Finetuning Text-to-Image Diffusion Models for Fairness

ICLR 2024arXiv

citations

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

ICLR 2025arXiv

citations

A Closer Look at Machine Unlearning for Large Language Models

ICLR 2025arXiv

citations

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

ICML 2025arXiv

citations

Exploring Incompatible Knowledge Transfer in Few-Shot Image Generation

CVPR 2023arXiv

citations

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift

CVPR 2025arXiv

citations

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

ICLR 2025arXiv

citations

Gaussian Mixture Solvers for Diffusion Models

NEURIPS 2023arXiv

citations

Locality Sensitive Sparse Encoding for Learning World Models Online

ICLR 2024arXiv

citations

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts

ICCV 2025arXiv

citations

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

NEURIPS 2025arXiv

citations

Improving Your Model Ranking on Chatbot Arena by Vote Rigging

ICML 2025arXiv

citations

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms

ICML 2025arXiv

citations

On Calibrating Diffusion Probabilistic Models

NEURIPS 2023arXiv

citations

Continual Reinforcement Learning by Planning with Online World Models

ICML 2025arXiv

citations

Chao Du

Affiliations

papers (22)

Understanding R1-Zero-Like Training: A Critical Perspective

On Evaluating Adversarial Robustness of Large Vision-Language Models

Efficient Diffusion Policies For Offline Reinforcement Learning

Scaling up Masked Diffusion Models on Text

Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast

When Attention Sink Emerges in Language Models: An Empirical View

Weak-to-Strong Jailbreaking on Large Language Models

Finetuning Text-to-Image Diffusion Models for Fairness

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

A Closer Look at Machine Unlearning for Large Language Models

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

Exploring Incompatible Knowledge Transfer in Few-Shot Image Generation

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

Gaussian Mixture Solvers for Diffusion Models

Locality Sensitive Sparse Encoding for Learning World Models Online

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

Improving Your Model Ranking on Chatbot Arena by Vote Rigging

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms

On Calibrating Diffusion Probabilistic Models

Continual Reinforcement Learning by Planning with Online World Models

papers (22)

Understanding R1-Zero-Like Training: A Critical Perspective

On Evaluating Adversarial Robustness of Large Vision-Language Models

Efficient Diffusion Policies For Offline Reinforcement Learning

Scaling up Masked Diffusion Models on Text

Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast

When Attention Sink Emerges in Language Models: An Empirical View

Weak-to-Strong Jailbreaking on Large Language Models

Finetuning Text-to-Image Diffusion Models for Fairness

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

A Closer Look at Machine Unlearning for Large Language Models

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

Exploring Incompatible Knowledge Transfer in Few-Shot Image Generation

Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

Gaussian Mixture Solvers for Diffusion Models

Locality Sensitive Sparse Encoding for Learning World Models Online

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

Improving Your Model Ranking on Chatbot Arena by Vote Rigging

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms

On Calibrating Diffusion Probabilistic Models

Continual Reinforcement Learning by Planning with Online World Models