About
Previously, I was a founding engineer at Contextual AI, working on agentic RAG, evaluation, and alignment mentored by Douwe Kiela. Before that, I was a researcher at Center for Brains Mind and Machines - MIT and Electronic visualization Lab - UIUC
Selected Publications
View All →Nikita Vassilyev, William Berrios, Ruowang Zhang, Bo Han, Douwe Kiela, Shikib Mehri
A unified framework treating context updates as an optimization problem, extending context learning with classical primitives like batching, failure replay, and grouped rollouts.
Jon Saad-Falcon†, Rajan Vivek†, William Berrios†, Nandita Shankar Naik, Matija Franklin, Bertie Vidgen, Amanpreet Singh, Douwe Kiela, Shikib Mehri
Findings of the Association for Computational Linguistics: EMNLP 2024
A natural language unit testing paradigm for fine-grained LLM evaluation, achieving state-of-the-art on FLASK and BigGenBench benchmarks.
Nicholas Lui, Bryan Chia, William Berrios, Candace Ross, Douwe Kiela
Proceedings of the AAAI Conference on Artificial Intelligence
Using diffusion model perturbations to create demographically balanced datasets for measuring racial bias in vision-language models.