Xin Liu
Xin Liu

About Me

Xin Liu (刘 馨) is a third-year master student at East China Normal University. Over the past year, her research has focused on Multi-Modal Large Language Model. Two papers were accepted by IJCAI 2024, ECCV 2024 respectively. One paper was submitted to NeurIPS 2024. She acts as the first author of these papers.

[Highlight] She is looking for a PhD position to start in 2025 Fall (not limited in China).

Interests
  • Multi-Modal Large Language Model
Education

  • MS Student

    East China Normal University

Featured Publications
Safety of Multimodal Large Language Models on Images and Text
MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models
Recent Publications
Xin Liu, Yichen Zhu, Yunshi Lan, Chao Yang, Yu Qiao (2024). Safety of Multimodal Large Language Models on Images and Text. IJCAI.
PDF Cite Code Talk
Xin Liu, Yichen Zhu (Equal First Author), Jindong Gu, Yunshi Lan, Chao Yang, Yu Qiao (2023). MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models. ECCV.
PDF Cite Code Project Page
Yunshi Lan, Xiang Li, Xin Liu, Yang Li, Wei Qin, Weining Qian (2023). Improving zero-shot visual question answering via large language models with reasoning question prompts. ACM MM.
PDF Cite
Yichen Zhu, Ning Liu, Zhiyuan Xu, Xin Liu, Weibin Meng, Louis Wang, Zhicai Ou, Jian Tang (2022). Teach less, learn more: On the undistillable classes in knowledge distillation. NeurIPS.
PDF Cite
Recent & Upcoming Talks
Safety of Multimodal Large Language Models on Images and Text
Recent News

✌ MM-SafetyBench is accepted by ECCV 2024!

As an early piece of work, this benchmark systematically reveals that current Multimodal Large Language Models are susceptible to malicious attacks.