About me

Hi, I am currently a Ph.D. student in the Health Intelligence (HI) group of the research center of Social Computing and Information Retrieval (SCIR) at the Harbin Institute of Technology under the supervision of Professor Bing Qin and Professor Sendong Zhao. My research interests lie in the safety and robustness of language models. In the era of small models, I focused on analyzing the spurious correlations captured by the model during training, which is one of the reasons for poor robustness. In the era of large language models (LLMs), I focused on making some applications (such as jailbreaking attacks and defense methods) and exploring safety or robustness issues in the transformation process from the general domain to the vertical domain.

Research Interests

• Robustness and Safety of LLMs
• AI + Pharmaceutical Science
• Biomedical LLMs

Education

Harbin Institute of Technology 2021-Now Ph.D. in Computer Science Supervisor: Professor Bing Qin and Associate Professor Sendong Zhao

Chinese University of Hong Kong 2024.11-Now Research Assistant, Department of Mathematics Supervisor: Professor Fenglei Fan

Northeastern University 2017-2021 Bachelor in Computer Science Supervisor: Associate Professor Feiliang Ren

Publications and Researches

• Less learn shortcut: Analyzing and mitigating learning of spurious feature-label correlation
  Yanrui Du, Jing Yan, Yan Chen, Jing Liu, Sendong Zhao, Qiaoqiao She, Hua Wu, Haifeng Wang, Bing Qin
  IJCAI 2023 [paper]
• Make Your Decision Convincing! A Unified Two-Stage Framework: Self-Attribution and Decision-Making
  Yanrui Du, Sendong Zhao, Haochun Wang, Yuhan Chen, Rui Bai, Zewen Qiang, Muzhen Cai, Bing Qin
  EMNLP 2023 Findings [paper]
• Don't Ignore Dual Logic Ability of LLMs while Privatizing: A Data-Intensive Analysis in Medical Domain
  Yanrui Du, Sendong Zhao, Muzhen Cai, Ming Ma, Danyang Zhao, Jiawei Cao, Bing Qin
  BIBM 2024 [paper]
• MoGU: A Framework for Enhancing Safety of LLMs While Preserving Their Usability
  Yanrui Du, Sendong Zhao, Danyang Zhao, Ming Ma, Yuhan Chen, Liangyu Huo, Qing Yang, Dongliang Xu, and Bing Qin
  NeurIPS 2024 [paper]
• Analyzing the Inherent Response Tendency of LLMs: Real-World Instructions-Driven Jailbreak
  Yanrui Du, Sendong Zhao, Ming Ma, Yuhan Chen, Bing Qin
  AAAI 2025 [paper]
• GLS-CSC: A Simple but Effective Strategy to Mitigate Chinese STM Models' Over-Reliance on Superficial Clue
  Yanrui Du, Sendong Zhao, Yuhan Chen, Rai Bai, Jing Liu, Hua Wu, Haifeng Wang, Bing Qin
  Arxiv Preprint [paper]

• From Artificially Real to Real: Leveraging Pseudo Data from Large Language Models for Low-Resource Molecule Discovery
  Yuhan Chen, Nuwa Xi, Yanrui Du, Haochun Wang, Chen Jianyu, Sendong Zhao, Bing Qin
  AAAI 2024 [paper]
• MolTailor: Tailoring Chemical Molecular Representation to Specific Tasks via Text Prompts
  Haoqiang Guo, Sendong Zhao, Haochun Wang, Yanrui Du, Bing Qin
  AAAI 2024 [paper]
• MolFusion: Multimodal Fusion Learning for Molecular Representations via Multi-granularity Views
  Muzhen Cai, Sendong Zhao, Haochun Wang, Yanrui Du, Zewen Qiang, Bing Qin, Ting Liu
  Arxiv Preprint [paper]

• BenTsao: Open-source Chinese Medical Large Language Model
  Haochun Wang, Yanrui Du, Chi Liu, Rui Bai, Nuwa Xi, Yuhan Chen, Zewen Qiang, Jianyu Chen, Zijian Li, Sendong Zhao, Bin Qin, Ting Liu
  Repo: https://github.com/SCIR-HI/Huatuo-Llama-Med-Chinese
• Knowledge-tuning Large Language Models with Structured Medical Knowledge Bases for Trustworthy Response Generation in Chinese
  Haochun Wang, Sendong Zhao, Zewen Qiang, Zijian Li, Nuwa Xi, Yanrui Du, MuZhen Cai, Haoqiang Guo, Yuhan Chen, Haoming Xu, Bing Qin, Ting Liu
  TKDD [paper]
• The CALLA Dataset: Probing LLMs’ Interactive Knowledge Acquisition from Chinese Medical Literature
  Yanrui Du, Sendong Zhao, Muzhen Cai, Jianyu Chen, Haochun Wang, Yuhan Chen, Haoqiang Guo, Bing Qin
  Arxiv Preprint [paper]

Projects

• Ministry of Science and Technology, CHINA - Human-machine Integration Consultation. 2021-2024.
• Huawei - Prompt Learning Research Project. 2022-2023.
• Du Xiaoman - Research on the correlation between decision-making and Chain-of-thought. 2023-2024.
• Baidu - Robustness Analysis of Language Models Project. 2021-2022.