Ellen Vitercik

Assistant Professor @Stanford Management Science & Engineering and Computer Science | Machine Learning & Optimization

[Rohan Sawhney @rohansawhney1]

Releasing Walk on Spheres Extensions (WoSX): a GPU-accelerated C++/Python library for Monte Carlo physics simulation on complex geometry

Think path tracing but for physics beyond light transport: heat, electrostatics, potential flow, deformation & more!

https://t.co/1F59rys9tw

[Ellen Vitercik @vitercik]

@thegautamkamath Congrats!! 👏

[Ellen Vitercik @vitercik]

Thanks to the organizers and participants at the Simons Institute for the fantastic workshop and discussion!

Paper: https://t.co/cb4l4M4aXB
Slides: https://t.co/vdgxlFJlUw
Code: https://t.co/rjY6qB7qvi
Data: https://t.co/RnfoZAtcuR
Workshop: https://t.co/sBsUZRk9eU

(9/9)

[Ellen Vitercik @vitercik]

Can we evaluate if LLMs can reason structurally?

I discussed this question through the lens of data structures in a talk at @SimonsInstitute. In a joint ICML'26 work with @yuhe441, Yingxi Li, and @crwhite_ml, we introduce DSR-Bench, the data structure reasoning benchmark. (1/9)

[Ellen Vitercik @vitercik]

On our Challenge subset, no model reaches 0.5 overall accuracy. (8/9)

[Ellen Vitercik @vitercik]

Slides: https://t.co/whVXsW7tvt

MAGNOLIA: Matching Algorithms via GNNs for Online Value-to-go Approximation: https://t.co/4FmTVottoa

Algorithms with Calibrated Machine Learning Predictions: https://t.co/NT4BqytzNY

[Ellen Vitercik @vitercik]

Can machine learning improve discrete optimization algorithms without sacrificing theoretical guarantees?

This was the central question of the talk I gave this spring at a few schools (UCSD, UIC, Yale, Penn): Machine Learning for Discrete Optimization: Theoretical Foundations.

[Ellen Vitercik @vitercik]

We investigate the value of calibration in online rent-or-buy and online job scheduling, with guarantees depending on accuracy, calibration error, sharpness, and inherent uncertainty.

Joint work with @judyhshen and Anders Wikum. ICML'25 Spotlight.

[Ellen Vitercik @vitercik]

Learn more and submit: https://t.co/BQiiN4neYK

[Ellen Vitercik @vitercik]

More time to submit to LAMP 2026! The deadline for spotlight talks, posters, and open problems has been extended to May 14.

Details in first reply.

[Ellen Vitercik @vitercik]

We’re excited to announce the call for submissions for our workshop on Learning-driven Algorithms and Machine-aided Proofs (LAMP) at Toyota Technical Institute of Chicago (@TTIC_Connect) on August 6–7.

We welcome spotlight talks, posters, and open problems at the intersection of machine learning and theoretical computer science. We have an exciting lineup of speakers from academia and industry, and we’re looking forward to bringing this emerging community together.

Huge thanks to my co-organizers Sandeep Silwal and Dravyansh Sharma.

Please submit your work or send us an open problem. Details in the first reply.