Hey there! I'm currently a research intern at CMU LTI working with Graham Neubig. I recently graduated from UC Berkeley 🐻, where I was very fortunate to be advised by Alane Suhr as part of the Berkeley NLP Group. Before that, I spent a great summer interning with the AI Technology Group at the MIT Lincoln Lab.

I'm drawn to ideas that develop a deep understanding of the underlying mechanisms, capabilities, and limitations of language models. Recently, I've been thinking about how we can develop reasoning models that are flexible and generalize well out-of-distribution.

In addition to research, I spent three fun semesters teaching the data structures & algos class (CS 61B) at Berkeley. In my free time, I enjoy being outdoors, playing video games with my friends, and sending funny cat videos to Angela.

Preprints & Publications

Long Chain-of-Thought Reasoning Across Languages
Josh Barua, Seun Eisape, Kayo Yin, Alane Suhr
SCALR @ COLM 2025
paper

Using Language Models to Disambiguate Lexical Choices in Translation
Josh Barua, Sanjay Subramanian, Kayo Yin, Alane Suhr
EMNLP 2024
paper| code

Improving Medical Visual Representations via Radiology Report Generation
Keegan Quigley, Miriam Cha, Josh Barua, Geeticka Chauhan, Seth Berkowitz, Steven Horng, Polina Golland
ArXiv
paper| cite