Robin Jia

Robin Jia is an Assistant Professor of Computer Science in the Thomas Lord Department of Computer Science at the University of Southern California, where he leads the AI, Language, Learning, Generalization, and Robustness (Allegro) Lab. He received his Ph.D. in Computer Science from Stanford University in 2020, advised by Percy Liang, and his Bachelor of Science with honors in Computer Science and a minor in Biology from Stanford University in 2014. Prior to his current role, Jia conducted postdoctoral research at Facebook AI Research, collaborating with Luke Zettlemoyer and Douwe Kiela. His professional experience also includes research internships at Microsoft Research Redmond focusing on biomedical machine reading in 2018, the Google Research Deep Dialogue team in 2016, the Google Crisis Response Team in 2014, and the YouTube Ads team in 2012.

Jia’s research specializes in natural language processing and machine learning, with emphasis on robustness, question answering, and large language models, including in-context learning, data memorization, numerical reasoning, factual recall, and AI policy issues such as data watermarking and fair learning frameworks for copyright claims. Notable publications include “Adversarial Examples for Evaluating Reading Comprehension Systems” (EMNLP 2017, Outstanding Paper Award), “Know What You Don’t Know: Unanswerable Questions for SQuAD” (ACL 2018, Best Short Paper Award), “Contrastive Novelty-Augmented Learning: Anticipating Outliers with Large Language Models” (ACL 2023, SoCalNLP Symposium 2022 Best Paper Award), “Transformers Learn Higher-Order Optimization Methods for In-Context Learning: A Study with Linear Models” (NeurIPS 2024, SoCalNLP Symposium 2023 Best Paper Award), and “Hubble: a Model Suite to Advance the Study of LLM Memorization” (ICLR 2026). He has received the Google ML and Systems Junior Faculty Award (2025), Google Research Scholar Award (2023), Cisco Research Award (2023), Joint NSF Grant (2024), and funding from USC-Amazon Center and USC-Capital One Center. Jia’s work advances benchmarking and evaluation of LLM capabilities and limitations.

Professional Email: robinjia@usc.edu