Publications

• MedGUIDE: Benchmarking Clinical Decision-Making in Large Language Models
  Xiaomin Li*, Mingye Gao*, Yuexing Hao, Taoran Li, Guangya Wan, Zihan Wang, Yijun Wang.
  NeurIPS GenAI4Health Workshop 2025

• SoK: Understanding (New) Security Issues Across AI4Code Use Cases
  Qilong Wu*, Taoran Li*, Tianyang Zhou*, Varun Chandrasekaran.
  Under review at Network and Distributed System Security Symposium 2026

• The Erasure Illusion: Stress-Testing the Generalization of LLM Forgetting Evaluation
  Hengrui Jia, Taoran Li, Jonas Guan, Varun Chandrasekaran.
  Under review at ACM Conference on Computer and Communications Security 2026

• Layer-Targeted Multilingual Knowledge Erasure in Large Language Models
  Taoran Li, Varun Chandrasekaran, Zhiyuan Yu.
  Under review at 43rd International Conference on Machine Learning (ICML 2026)

• CounterFlow: Securing LLM Agents via Graph Counterfactuals
  Taoran Li, Zhiyuan Yu.
  Under review at the 33rd ACM Conference on Computer and Communications Security (CCS 2026)

• Chain of Risk: Safety Failures in Large Reasoning Models and Mitigation via Adaptive Multi-Principle Steering
  Xiaomin Li, Jianheng Hou, Zheyuan Deng, Zhiwei Zhang, Taoran Li, Binghang Lu, Bing Hu, Yunhan Zhao, Yuexing Hao.
  Under review at the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS 2026)

• Same Answer, Different Reasons: Trace-Level Divergence in Multilingual Large Language Models
  Rui Sun, Peizhao Mei, Xiwei Cheng, Kaizhuo Chen, Taoran Li, Zhiyuan Yu, Varun Chandrasekaran.
  Under review at the Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS 2026)

• AutoDojo: Adaptive Attacks Expose Superficial Defenses and User-Underspecification Limits in LLM Agents
  Xinhang Ma, Taoran Li, Chaowei Xiao, Zhiyuan Yu, Ning Zhang, Yevgeniy Vorobeychik.
  Under review at the 48th IEEE Symposium on Security and Privacy (S&P 2027)