Scholar
Yichi Zhang
Google Scholar ID: HzgDakoAAAAJ
PhD candidate, Tsinghua University
Citations & Impact
Citations
834
H-index
10
i10-index
11
Publications
20
Co-authors
8
list available
Publications
Resume (English only)
Academic Achievements
- Two papers accepted at NeurIPS 2025
- Oral presentation at ICML 2025: 'STAIR: Safe alignment with introspective reasoning'
- Highlight paper at CVPR 2024
- Paper accepted at ICLR 2024
- Paper accepted at CVPR 2023
- Paper published in CVIU
- Released RealSafe-R1 (2024), a safety-aligned DeepSeek-R1 family without compromising reasoning
- Released MultiTrust benchmark (2024) for trustworthiness evaluation of multimodal LLMs
- 1st place in 2022 International Algorithm Case Competition (Adversarial Robustness track)
- Selected for Tencent 'Hunyuan Scholar' PhD Research Incentive Program (2025, 23 nationwide)
- Beijing Outstanding Graduate (2022, only 4 in CS department)
- Merit Student, Beijing (2021)
- 2nd place in CVPR 2021 Security AI Challenger (Unrestricted Adversarial Attacks on ImageNet, 2/1599)
- 8th place in CIKM 2020 Adversarial Challenge on Object Detection (8/1814)
Research Experience
- Research Intern, Qwen Team, Alibaba, 2025–Present
- Visiting Ph.D. Student, TML Lab (Prof. Tongliang Liu), University of Sydney, Spring 2025
- Research Intern, RealAI, 2023–2025
- Research Intern, Tencent, 2021
- Closely collaborates with Prof. Hang Su and Prof. Yinpeng Dong
