钱意(Yi Qian)

钱意(Yi Qian)

Ph.D Candidate in Nanjing University

About me

Currently, I am a Ph.D. candidate at Nanjing University, affiliated with the National Key Laboratory for Novel Software Technology. I work in the System and Software Security Laboratory under the supervision of Professor Bing Mao.

I received my B.E. degree in information security from the College of Computer Science, Chongqing University, China, in 2021. In the same year, I was admitted to the Ph.D. program at Nanjing University.

Random quote

Ciallo～(∠・ω< )⌒☆

Research Interest

My research centers on the intersection of security for AI and AI for security.

Security for AI

  • Agent–environment interactions. Characterizing security boundaries across language models, tools, and software execution environments.
  • Agent and harness vulnerabilities. Investigating how context construction, instruction hierarchies, and tool-mediated execution expose agents to adversarial manipulation.
  • Security evaluation and defense. Developing systematic analyses, evaluation methods, and defenses for AI-driven applications.

AI for Security

  • Program understanding and specification inference. Applying language models to recover program semantics and infer specifications for security analysis.
  • Vulnerability discovery and validation. Integrating model-guided reasoning with program analysis and testing to identify and validate software vulnerabilities.
  • Automated exploitability assessment. Generating proof-of-concept tests and exploit prototypes to reproduce vulnerabilities and assess their security impact.

Publications

  1. Qian, Y., Qian, K., He, X., Chen, L., Zhang, J., Zhang, T., Wei, H., Wang, L., Wu, H., & Mao, B. (2026, November). Mind the Gap: Action Rebinding Attacks against Android GUI Agents. In The 33rd ACM Conference on Computer and Communications Security (CCS). (CCF-A Conference)

  2. He, X., Chen, Y., Qian, Y., Wei, H., Chen, L., Fu, Z., Wang, L., Wu, H., & Mao, B. (2026, August). When Context Gets Root: Privilege Escalation in LLM Harnesses. arXiv preprint arXiv:2608.27299.

  3. Qian, Y., Peng, F., Wu, H., Chen, L., & Mao, B. (2025, November). Uncovering Prompt Elements: Cloning System Prompts from Behavioral Traces. In 2025 40th IEEE/ACM International Conference on Automated Software Engineering (ASE) (pp. 457-468). IEEE. (CCF-A Conference)

  4. Chen, L., He, Z., Qian, Y., & Mao, B. (2025). CATI++: Empirical Study and Evaluation for Adjacent Instruction Enhanced Type Inference. The Computer Journal, bxaf004. (CCF-B Journal)

  5. Chen, L., Qian, Y., Wang, Y., & Mao, B. (2023). Nimbus++: Revisiting Efficient Function Signature Recovery with Depth Data Analysis. International Journal of Software Engineering and Knowledge Engineering, 33(10), 1537-1565. (CCF-C Journal)

  6. Qian, Y., Chen, L., Wang, Y., & Mao, B. (2022, December). Nimbus: Toward Speed Up Function Signature Recovery via Input Resizing and Multi-Task Learning. In 2022 IEEE 22nd International Conference on Software Quality, Reliability and Security (QRS) (pp. 454-463). IEEE. (CCF-C Conference)

  7. Chen, L., He, Z., Wu, H., Xu, F., Qian, Y., & Mao, B. (2022, March). Dicomp: Lightweight Data-Driven Inference of Binary Compiler Provenance with High Accuracy. In 2022 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER) (pp. 112-122). IEEE. (CCF-B Conference)

Honors & Awards

  • 2025Outstanding Graduate Student, Nanjing University
  • 2025Graduate Talent Scholarship
  • 2020Second Prize, National College Student Information Security Competition (Project Track); project Antideepfake: Countering AI Face-Swapped Videos
  • 2017–2019First-Class Scholarship (three times)
  • 2017–2018Outstanding Student, Chongqing University