Peiru Yang

Peiru Yang

Ph.D. Candidate at Department of Electronic Engineering, Tsinghua University.

Peiru Yang (杨珮茹)

I am currently a Ph.D. Candidate in the Department of Electronic Engineering at Tsinghua University, advised by Prof. Yongfeng Huang, and I also work closely with Prof. Tao Qi from Beijing University of Posts and Telecommunications. Before that, I received my B.E. degree from Tsinghua University in 2021.

My research interests primarily lie in LLM pretrain data strategy and LLM safety.

📖 Selected Publications

(For a full list, please refer to my Google Scholar)

  • Yang, P., Luo, Y., Gao, Z., Ju, T., Zheng, H., Zhu, L., Fu, H., Li, Q., Wang, S., & Qi, T. (2026). Towards Whole-corpus Reconstruction of Heterogeneous RAG Knowledge Bases. International Conference on Machine Learning (ICML 2026).

  • Yang, P., Zheng, H., Ju, T., Wang, S., Ni, W., Liu, J., Wang, S., Huang, Y., & Qi, T. (2026). Knowledge Poisoning Attacks on Medical Multi-Modal Retrieval-Augmented Generation. Annual Meeting of the Association for Computational Linguistics (ACL 2026).

  • Yang, P., Li, Y., Wang, S., Liu, X., Gan, H., Li, X., Gao, Q., & Huang, Y. (2026). OncoCoT: A Temporal-causal Chain-of-Thought Dataset for Oncologic Decision-Making. AAAI Conference on Artificial Intelligence (AAAI 2026) (Oral). [Link]

  • Yang, P., Yin, J., Zheng, H., Bai, X., Wang, H., Sun, Y., Li, X., Pei, S., Huang, Y., & Qi, T. (2026). MrM: Black-Box Membership Inference Attacks against Multimodal RAG Systems. AAAI Conference on Artificial Intelligence (AAAI 2026) (Oral). [Link]

  • Yang, P., Zheng, H., Luo, Y., Liu, X., Wang, J., Wang, H., Li, X., Huang, Y., & Qi, T. (2026). ShieldRAG: Safeguarding Retrieval-Augmented Generation from Untrusted Knowledge Bases. AAAI Conference on Artificial Intelligence (AAAI 2026). [Link]

  • Yin, J., Yang, P. (co-first author), Yang, C., Wang, H., Hu, Z., Wang, S., Huang, Y., & Qi, T. (2025). Black-Box Membership Inference Attack for LVLMs via Prior Knowledge-Calibrated Memory Probing. Conference on Neural Information Processing Systems (NeurIPS 2025). [Link]

  • Yang, P., Wang, H., Huang, Y., Yang, S., Zhang, Y., Huang, L., Zhang, Y., Wang, G., Yang, S., He, L., & Huang, Y. (2024). LMKG: A Large-Scale and Multi-Source Medical Knowledge Graph for Intelligent Medicine Applications. Knowledge-Based Systems (KBS). [Link]

🎓 Education

  • Ph.D. in Electronic Engineering, Tsinghua University, Beijing, Sept. 2021 - Present. (Award: Future Scholar Scholarship)
  • B.E. in Electronic Engineering, Tsinghua University, Beijing, Sept. 2017 - June 2021. (GPA: 3.89/4.0, Rank 8/261; Award: Excellent Undergraduate Thesis)
  • Exchange Student, Computer Science, Rice University, Houston, Aug. 2019 - Jan. 2020. (GPA: 4.0/4.0)

💻 Work Experience

  • Research Intern, Xiaohongshu Hilab, Mar. 2026 - Present. Dots foundation model pretraining; Focused on pretraining data strategy and scaling law exploration for efficient training.
  • Research Intern, JD Health, Nov. 2023 - Aug. 2024. Focused on Medical Embedding Model Optimization and Heterogeneous RAG Frameworks; Deployed in internal systems.
  • Research Intern, Microsoft Research Asia (MSRA), Dec. 2020 - June 2021. Social Computing Group; Worked on Personalized News Recommendation and Bing News/Ads optimization.