Posts by Collection

portfolio

publications

Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification

Published in ICML 2025, 2025

Recommended citation: Yuhao Sun*, Jiacheng Zhang*, Zesheng Ye, Chaowei Xiao, Feng Liu. (2025). "Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification." ICML 2025. (*Equal contribution)

[Paper] [Code]

Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!

Published in ICLR 2026, 2026

Recommended citation: Zhexin Zhang, Yuhao Sun, Junxiao Yang, Shiyao Cui, Hongning Wang, Minlie Huang. (2026). "Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!" ICLR 2026.

[Paper] [Code]

Let’s Roll a BiFTA: Bi-refinement for Fine-grained Text-visual Alignment in Vision-Language Models

Published in TMLR 2026, 2026

Recommended citation: Yuhao Sun, Chengyi Cai, Jiacheng Zhang, Zesheng Ye, Xingliang Yuan, Feng Liu. (2026). "Let's Roll a BiFTA: Bi-refinement for Fine-grained Text-visual Alignment in Vision-Language Models." TMLR 2026.

[Paper] [Code]

From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents

Published in arXiv preprint, 2026

Recommended citation: Yuhao Sun, Jiacheng Zhang, Shaanan Cohney, Zhexin Zhang, Feng Liu, Xingliang Yuan. (2026). "From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents." arXiv preprint arXiv:2606.05805.

[Paper] [Code] [Project Page]

Yuhao Sun

Posts by Collection

portfolio

publications

Sample-Specific Noise Injection For Diffusion-Based Adversarial Purification

Be Careful When Fine-tuning On Open-Source LLMs: Your Fine-tuning Data Could Be Secretly Stolen!

Let’s Roll a BiFTA: Bi-refinement for Fine-grained Text-visual Alignment in Vision-Language Models

From Risk Classification to Action Plan Remediation: A Guardrail Feedback Driven Framework for LLM Agents

talks

teaching