Victor Ion Butoi

PhD Student
Massachusetts Institute of Technology
EECS, CSAIL
vbutoi@mit.edu

Greetings! I am a 1st 2nd 3rd year Computer Science PhD student at MIT fortunate enough to be advised by Professors Adrian Dalca and John Guttag as a part of the Clinical and Applied Machine Learning (CAML) Group. I am interested in problems lying at the intersection of machine learning, computer vision, and statistical inference, often with applications to healthcare. My work is supported by the NSF Graduate Research Fellowship.

Previously, I studied Computer Science at Cornell University advised by Mert Sabuncu, where I was a Merril Presidential Scholar for the Computer Science department and a Tanner-Dean Scholar for the School of Arts and Sciences.

My current focus is on solving traditional machine learning problems by re-thinking neural network inference. Here are some directions I am excited about:

  • Confidence Calibration: How can extend theoretical insights from statistics, such as calibration and proper scoring rules, derived in simple (x,y) classification settings, to a broad class of high-dimensional and structured machine learning problems?
  • In-Context Adaptation: How can we improve (and understand) neural networks' ability to meta-solve tasks at inference, without additional training?

If you are interested in collaborating, please do reach out!

Selected Publications

Image 3
UniverSeg: Universal Medical Image Segmentation
Victor Ion Butoi*, Jose J. Ortiz*, Tianyu Ma, John Guttag,
Mert R. Sabuncu, Adrian V. Dalca

ICCV 2023, MedNeurIPS 2022 (NeurIPS Workshop)

[Paper] [Code] [Project Page]

All Publications

Image 3
ConMe: Rethinking Evaluation of Compositional Reasoning for Modern VLMs
Irene Huang, Wei Lin, Muhammad Jehanzeb Mirza, Jacob A Hansen, Sivan Doveh, Victor Ion Butoi, Roei Herzig,
Assaf Arbelle, Hilde Kuehne, Trevor Darrell, Chuang Gan, Aude Oliva, Rogerio Feris, Leonid Karlinsky

NeurIPS 2024 (Datasets and Benchmarks Track)

[Paper] [Code]
Image 3
DEUP: Direct Epistemic Uncertainty Prediction
Moksh Jain*, Salem Lahlou*, Hadi Nekoei, Victor Ion Butoi, Paul Bertin, Jarrid Rector-Brooks,
Maksym Korablyov, Yoshua Bengio

TMLR 2023

[Paper] [Code]

Workshop Papers

Image 3
Generative Active Learning for the Search of Small-molecule Protein Binders
Maksym Korablyov et al.

ICLR GEM 2024 (Oral Presentation)

[Paper]
Image 3
Empirical Analysis of a Segmentation Foundation Model in Prostate Imaging
Heejong Kim, Victor Ion Butoi, Mert R. Sabuncu, Adrian V. Dalca

MICCAI MedAGI 2023

[Paper]

Reviewer Service

AutoML, MIDL, ICLR, ICML, NeurIPS

Bio

Previously, I have been lucky enough to work with: