Sophia Kolak

PhD student @ CMU's S3D, advised by Claire Le Goues, Vincent Hellendoorn, and Ruben Martins, where I study LLMs fine-tuned on source code and automatic program repair. I also worked at IBM Quantum, studying how quantum algorithms for graph partioning could be used to improve linear solvers. Prior to that, I was a student at Columbia University, where I worked in the ARiSE lab using machine learning to predict the efficiency of programs.

   email: sophiakolak@cmu.edu  cv: pdf

Publications

Revisiting Unnaturalness for Automated Program Repair in the Era of Large Language Models
Aidan Yang, Sophia Kolak, Vincent J. Hellendoorn, Ruben Martins, Claire Le Goues
ICSE 2025 preprint

Evaluating Quantum Algorithms For Linear Solver Workflows
Sophia Kolak, Hamed Mohamedbagherpoor, Konstantis Daloukas, Kostas Kafousas, Francois-Henry Rouet, Yorgos Koutsoyannopoulos, Nathan Earnest-Noble, Robert Lucas
International Supercomputing Conference, 2023 link

Patch Generation with Language Models: Feasibility and Scaling Behavior
Sophia Kolak, Ruben Martins, Claire Le Goues, Vincent Hellendoorn
ICLR DL4C workshop 2022 link

It Takes a Village to Build a Robot: An Empirical Study of the ROS Ecosystem
Sophia Kolak, Afsoon Afzal, Claire Le Goues, Michael Hilton, Chris Timperley
ICSME 2020 link

Detecting Performance Patterns with Deep Learning
Sophia Kolak
SPLASH companion 2020 link

SHIRLEE: A Sharp-edge Handheld Identifier and Remover in Low-gravity Extravehicular Environments
Dada, Ganeshan, Groll, Kolak, Ravi, Stein, Wang
AIAA SciTech Forum 2021 link

Presentations

Quantum Algorithms for Linear Solvers
International Supercomputing Conference, Hamburg 2023  Slides

Patch Generation with Large Language Models
ICLR DL4C Workshop, virtual 2022  Poster

Robotics Software Quality Panelist
ROS World, Virtual 2020  Video

ROS Developers Podcast
Featured Interview 2020  Video

It Takes a Village to Build a Robot
ROScon, Macau 2019  Video Slides

Teaching

MLG 10615: Art & Machine Learning

Carnegie Mellon University, Spring 2023

COMS W3261: Computer Science Theory

Columbia University, Fall 2019-Spring 2021