My work bridges natural language processing and human–AI interaction. At its core, I am driven by two questions: How can machines grasp the social and cultural layers of human communication, and how can AI be harnessed to enrich human experiences across diverse domains? To this end, I explore how computational systems can engage with the quirks of language—figurative turns, idioms, and euphemisms—that make communication so uniquely human. This line of inquiry not only exposes the technical challenges of reasoning over language but also reveals the deeper patterns of how people express themselves. A parallel strand of my research translates these insights into practical advances, creating novel educational tools that support STEM learning and medical training.

Prior to coming to Princeton, I was an Assistant Professor in Electrical and Computer Engineering at the University of Illinois at Urbana-Champaign. I was also an affiliate faculty member in the departments of Computer Science, Educational Psychology, the National Center for Supercomputing Applications and the Center for Social and Behavioral Science, which permitted me to be engaged in interdisciplinary research.

Publications

For a full list of my publications, visit my Google Scholar page.

A sample of my recent publications:

2025

• Medical students’ perception of automated note feedback after simulated encounters

  Co-authored with S. Bansal, M.J. Yadav, and others

  [The Clinical Teacher]

• An LLM-Based Framework for Simulating, Classifying, and Correcting Students’ Programming Knowledge with the SOLO Taxonomy

  S. Zhang, P. S. Meshram, P. Ganapathy Prasad, M. Israel, and S. Bhat

  [SIGCSE 2025]

2024

• Intermediate Fine-Tuning Improves Mathematical Reasoning in Smaller Models

  N. Gangwar, S. Bhat, and N. Kani

  [Workshop on Mathematical Reasoning and AI at NeurIPS’24]

• Enhancing Language Models with Idiomatic Reasoning

  J. Zhou, Z. Zeng, H. Gong, and S. Bhat

  [COLM 2024]

• CLASP: Cross-modal Alignment Using Pre-trained Unimodal Models

  J. Zhou, Z. Zeng, H. Gong, and S. Bhat

  [Findings of ACL 2024]