Yangfeng Ji

Yangfeng Ji is an Associate Professor in the Department of Computer Science at the University of Virginia, where he has been since 2018. Previously, he held a postdoctoral position at the Paul G. Allen School of Computer Science & Engineering, University of Washington, from 2016 to 2018. He earned his PhD in Computer Science from the Georgia Institute of Technology in 2016. His research focuses on Natural Language Processing (NLP) and Machine Learning, with an emphasis on ethical AI, bias mitigation, and model interpretability.

Education:

• PhD in Computer Science, Georgia Institute of Technology, 2016
• Postdoctoral Researcher, Paul G. Allen School of Computer Science & Engineering, University of Washington, 2016–2018

Research Interests:
Dr. Ji’s work addresses critical challenges in NLP and Machine Learning, including fairness and bias in Large Language Models (LLMs), model interpretability through techniques like saliency estimation and rationale evaluation, and the development of robust evaluation frameworks. His contributions span topics such as gender representation vectors in LLMs, allocational harms, and improving temporal awareness in recommendation systems. He also explores data selection methods for model fine-tuning and secure data appraisal techniques.

Recent Research Trends:
His publications emphasize addressing biases and fairness in LLMs, enhancing model interpretability via contrastive activation analysis, and improving robustness through adversarial training and data selection. Key themes include mitigating vulnerabilities in model explanations and optimizing latent spaces for out-of-distribution detection.

Scientific Awards:
No awards explicitly mentioned in the provided information.

Advising and Grants:
No advisees or grant details provided in the text. Dr. Ji’s research group focuses on collaborative projects in ethical AI and NLP applications.

Labs/Teams:
No specific lab or team affiliations explicitly stated in the text.