Tanqiu Jiang | Stony Brook University

About Me

Hi, welcome to my personal website!
I am a Ph.D. student at the Stony Brook University in the Department of Computer Science where I intend to study the reliability issues of LLMs and the Privacy and Security issues of AI models. I am advised by Dr. Ting Wang and received one year of kind advise and help from Dr. Dongwon Lee at Penn State University.

Research Interests

Reliable LLMs: Jailbreaking, Hallucination, KV Cache Optimization, Retrieval-Augmented Generation
Security & Privacy: Adversarial Machine Learning, Unlearnable Examples, Differentially Private Models

News

[Jan. 2025] Our paper about defending jailbreak attacks via attention-based KV eviction is accepted to ICLR 2025! See you in Singapore!
[Jan. 2025] Our paper about Retrieval-Augmented Differentially Private Diffusion Model is accepted to ICLR 2025! See you in Singapore!
[Aug. 2023] I have started a new journey at Stony Brook University as a PhD Student in Computer Science!

Selected Publications

ICLR

RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction

Tanqiu Jiang, Zian Wang, Jiacheng Liang, Changjiang Li, Yuhui Wang, Ting Wang

International Conference on Learning Representations (ICLR), 2025.

PDF Accept (poster)

ICLR

RAPID: Retrieval Augmented Training of Differentially Private Diffusion Models

Tanqiu Jiang, Changjiang Li, Fenglong Ma, Ting Wang

International Conference on Learning Representations (ICLR), 2025.

PDF Code Accept (poster)

ICLR

Learning-Augmented Data Stream Algorithms

Tanqiu Jiang, Yi Li, Honghao Lin, Yisong Ruan, David P. Woodruff

International Conference on Learning Representations (ICLR), 2020.

PDF Code Accept (poster)

Services

Conference Reviewers

International Joint Conference on Neural Networks (IJCNN) 2024