Scholar
Jingyang Zhang
Google Scholar ID: f3DQwmgAAAAJ
ECE Ph.D. @ Duke University
Citations & Impact
Citations
833
H-index
14
i10-index
14
Publications
20
Co-authors
4
list available
Publications
Resume (English only)
Academic Achievements
- - Publications:
- - DVERGE: Diversifying Vulnerabilities for Enhanced Robust Generation of Ensembles, NeurIPS 2020
- - Mixture Outlier Exposure: Towards Out-of-Distribution Detection in Fine-Grained Environments, WACV 2023
- - Openood v1.5: Enhanced benchmark for out-of-distribution detection, arXiv preprint, 2023
- - Awards: Min-K%++ method accepted to ICLR 2025 as Spotlight paper
- - Open-source projects: OpenOOD, lmms-finetune, VLM-Visualizer
Research Experience
- - Research Scientist at Virtue AI, focusing on LLM safety (Aug 2025 - present)
- - Machine Learning Engineer at Sciforium, developing multi-modal LLMs (Jan 6, 2025 - Aug 2025)
- - Machine Learning Intern at Tesla (May 13, 2023 - Summer 2023)
- - Internship experience at Bosch Center for AI
Education
- - Ph.D.: 2024, Duke University, Department of Electrical and Computer Engineering, Advisors: Prof. Yiran Chen and Prof. Hai (Helen) Li
- - B.Eng.: 2019, Tsinghua University, Department of Electronic Engineering
Background
- - Research Interests: LLM safety (redteaming, guardrail, etc.)
- - Professional Field: Adversarial machine learning and distributional shifts (specifically OOD detection)
- - Brief Introduction: Currently a Research Scientist at Virtue AI, focusing on LLM safety.
Miscellany
- - Personal Interests: Open-source projects
- - Technical Blogs: Medium and Zhihu
