Scholar
Zhengwei Fang
Google Scholar ID: v5lIRxUAAAAJ
Beijing Jiaotong University
Citations & Impact
Citations
284
H-index
5
i10-index
4
Publications
10
Co-authors
0
Contact
Publications
Resume
Academic Achievements
- Published multiple papers, including one highlighted at CVPR 2024 titled 'ANDA: Strong Transferable Adversarial Attacks via Ensembled Asymptotically Normal Distribution Learning'; developed the TransferAttack framework to boost adversarial transferability for image classification.
Research Experience
- Involved in several research projects on trustworthy AI, including MMTrustEval (a toolbox for benchmarking the trustworthiness of multimodal large language models) and MLA-Trust (a toolbox for benchmarking the trustworthiness of Multimodal LLM Agents across dimensions of truthfulness, controllability, safety, and privacy). Also worked on strong transferable adversarial attacks.
Background
- Research interests include, but are not limited to, trustworthy artificial intelligence.
Miscellany
- Lives in Beijing, China.