I build generative models, with a focus on diffusion models and language model pretraining.

I'm a PhD candidate at Cornell, co-advised by Carla P. Gomes and Kilian Q. Weinberger. My recent work covers scaling laws for data-constrained training, diffusion-augmented language modeling, and guided diffusion sampling. In 2025 I was a Research Scientist Intern at Meta, and before my PhD I spent three years in industry leading DARPA-funded AI research.

Research areas

  • Diffusion models
  • LLM pretraining
  • Scaling laws
  • Inverse problems
  • Efficient evaluation

Publications

* Equal contribution. Full list on Google Scholar.

Experience

  1. 2025

    Research Scientist Intern · Meta

    Collaborated with FAIR and Reality Labs on fine-tuning language models with memory layers specialized to specific domains.

  2. 2021 – now

    Graduate Researcher · Cornell University

    Generative modeling research across language model pretraining, scaling laws, and diffusion for scientific inverse problems.

  3. 2018 – 2021

    AI Developer · Stottler Henke Associates

    Led a DARPA-funded pathogen classification project through Phase I and Phase II, managing a small team. Built ML systems for NASA and the Air Force Research Laboratory, and won research grants from DARPA, NASA, and NGA.

Service & Teaching

Reviewer
ICLR 2027
NeurIPS 2026
ICML 2026 (Gold Reviewer)
CVPR 2026
ICLR 2026
NeurIPS 2025
Teaching Assistant
Fall 2022CS 6784 Advanced Topics in Machine Learning: Deep Learning
Spring 2022CS 4780 Introduction to Machine Learning
Fall 2021CS 4410 Operating Systems
Award
2022Cornell CIS Teaching Assistant Award