Portrait of Gustaf Ahdritz

I’m a fifth-year PhD student in Computer Science at Harvard University. I’m a member of the Machine Learning Foundations Group, advised by Boaz Barak, and supported by a graduate fellowship from Harvard’s Kempner Institute.

My primary research interest at the moment is uncertainty quantification, especially for large language models in realistic, unconstrained settings. Highlights include infrasampling, a framework for sampling from language models without hallucinations by abstaining where they are epistemically uncertain; a new approach for automatically generating uncertainty labels in unconstrained text at scale; and a rigorous framework for evaluating and improving uncertainty decompositions. Along the way, I've also worked on enabling real-time interactivity in LLMs and a benchmark measuring their ability to screen out untrustworthy information.

Previously, I interned at Meta in 2025 and Apple in 2024. I earned a B.A. in Computer Science and History (2020) and an M.S. in Computer Science (2021) from Columbia, where I worked with Mohammed AlQuraishi on protein structure prediction and led the development of OpenFold.

All on Google Scholar ↗
Human and model annotations collaborating on a manuscript
NeurIPS Position Paper Track · Spotlight

Large language models can not and should not be banned from peer review

Anat Kleiman*, Gustaf Ahdritz*

Bans on LLM-assisted reviewing are unenforceable. Since frontier models already rival human reviewers at identifying a paper's strengths and weaknesses, conferences should officially permit LLM assistance and make sure reviewers use the best models available.

Earlier work 3 publications

* Equal contribution.

Teaching

2023
Foundations of Deep Learning
Harvard COMPSCI 229br
2019–21
Advanced Programming
Columbia COMS 3157

Awards & fellowships

2020
Kosoresow Memorial Award
2019
Dean Hawkes Memorial Prize