
UBC Computer Science shines at annual international machine learning conference
Twenty-one papers from UBC Computer Science were accepted at ICML 2025 and its accompanying workshops
Thousands of researchers from around the world gathered in Vancouver from July 13 - 19, 2025, to discuss cutting-edge research in machine learning, artificial intelligence, statistics and data science at the 42nd annual International Conference on Machine Learning (ICML).
Researchers from UBC’s Department of Computer Science published 11 papers at the main conference this year. These papers included training large language models to solve complicated math problems, optimizing decision trees to be both accurate and scalable, and creating algorithms to analyze remote sensing data.
Two of the eleven papers are “Spotlight” papers, considered to be noteworthy and impactful in their field and are in the top ~6% of accepted papers at the conference. The two Spotlight papers from UBC Computer Science include one from Dr. Frank Wood’s group and one from Dr. Margo Seltzer’s group. Furthermore, the Spotlight paper from Dr. Seltzer’s group was chosen for an oral presentation, representing the top ~3% of all conference papers.
The following list of papers from UBC computer science researchers were accepted at this year’s conference:
- BEST-Route: Adaptive LLM Routing with Test-Time Optimal Compute
Dujian Ding, Ankur Mallick, Shaokun Zhang, Chi Wang, Daniel Madrigal, Mirian Hipolito Garcia, Menglin Xia, Laks Lakshmanan, Qingyun Wu, Victor Ruehle
Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret
Bingshan Hu, Zhiming Huang, Tianyue Zhang, Mathias Lécuyer, Nidhi Hegde
Direct Motion Models for Assessing Generated Videos
Kelsey Allen*, Carl Doersch, Guangyao Zhou, Mohammed Suhail, Danny Driess, Ignacio Rocco, Yulia Rubanova, Thomas Kipf, Mehdi S. M. Sajjadi, Kevin Murphy, Joao Carreira, Sjoerd van Steenkiste*
Galileo: Learning Global & Local Features of Many Remote Sensing Modalities
Gabriel Tseng*, Anthony Fuller*, Marlena Reil, Henry Herzog, Patrick Beukema, Favyen Bastani, James Green, Evan Shelhamer, Hannah Kerner, David Rolnick
Leveraging Online Olympiad-Level Math Problems for LLMs Training and Contamination-Resistant Evaluation
Sadegh Mahdavi*, Muchen Li, Kaiwen Liu, Christos Thrampoulidis, Leonid Sigal, Renjie Liao
Leveraging Predictive Equivalence in Decision Trees
Hayden McTavish*, Zachery Boner*, Jon Donnelly*, Margo Seltzer, Cynthia Rudin
Near-Optimal Decision Trees in a SPLIT Second
Varun Babbar*, Hayden McTavish*, Cynthia Rudin, Margo Seltzer
Spotlight paper and oral presentation
Near-optimal Sketchy Natural Gradients for Physics-Informed Neural Networks
Maricela Best Mckay, Avleen Kaur, Chen Greif, Brian Wetton
Position: When Incentives Backfire, Data Stops Being Human
Sebastin Santy, Prasanta Bhattacharya, Manoel Ribeiro, Kelsey Allen, Sewoong Oh
Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks
Yixin Cheng, Hongcheng Guo, Yangming Li, Leonid Sigal
Towards a Mechanistic Explanation of Diffusion Model Generalization
Matthew Niedoba, Berend Zwartsenberg, Kevin Murphy, Frank Wood
Spotlight paper
Two UBC computer science professors had leadership roles in helping to organize the conference this year: Dr. Kevin Leyton-Brown served as one of the social chairs on the organizing committee while Dr. Evan Shelhamer helped organize various social events as well as the 2nd Workshop on Test-Time Adaptation: Putting Updates to the Test (PUT) workshop. Dr. Jeff Clune was an invited speaker at the Exploration in AI Today workshop and Dr. Shelhamer was an invited speaker at the Championing Open-source Development in Machine Learning (CODEML) workshop.
Ten more papers were accepted at ICML workshops:
Adaptive Diffusion Denoised Smoothing : Certified Robustness via Randomized Smoothing with Differentially Private Guided Denoising Diffusion
Frederick Shpilevskiy, Saiyue Lyu, Krishnamurthy Dj Dvijotham, Mathias Lécuyer, Pierre-Andre Noel
2nd Workshop on Test-Time Adaptation: Putting Updates to the Test! (PUT)
EpitopeGen: Learning to Generate T Cell Epitopes: A Semi-Supervised Approach with Biological Constraints
Minuk Ma, Wilson Tu, Carlos Vasquez-Rios, Jiarui Ding
2nd Generative AI for Biology Workshop
GeoCrossBench: Cross-Band Generalization for Remote Sensing
Hakob Tamazyan, Ani Vanyan, Alvard Barseghyan, Anna Khosrovyan, Evan Shelhamer, Hrant Khachatrian
TerraBytes: Towards global datasets and models for Earth Observation
Implicit Bias of Spectral Descent and Muon on Multiclass Separable Data
Chen Fan, Mark Schmidt, Christos Thrampoulidis
3rd Workshop on High-dimensional Learning Dynamics (HiLD)
Learning What Matters: Prioritized Concept Learning via Relative Error-driven Sample Selection
Shivam Chandhok*, Qian Yang*, Oscar Mañas, Kanishk Jain, Aishwarya Agrawal, Leonid Sigal
ES-FoMo III: 3rd Workshop on Efficient Systems for Foundation Models
Less is More? Data Specialization for Self-Supervised Remote Sensing Models
Alvard Barseghyan*, Ani Vanyan*, Hakob Tamazyan, Evan Shelhamer, Hrant Khachatrian
TerraBytes: Towards global datasets and models for Earth Observation
On the Effect of Negative Gradient in Group Relative Deep Reinforcement Optimization
Wenlong Deng, Yi Ren, Muchen Li, Danica J. Sutherland, Xiaoxiao Li, Christos Thrampoulidis
2nd AI for Math Workshop
On the Performance of Differentially Private Optimization with Heavy-Tail Class Imbalance
Qiaoyue Tang, Alain Zhiyanov, Mathias Lécuyer
3rd Workshop on High-dimensional Learning Dynamics (HiLD)
Provenance Design and Evolution in a Production ML Library
Adam C Pocock, Joseph Wonsil, Romina Mahinpei, Jack Sullivan, Margo Seltzer
CODEML: Championing Open-source DEvelopment in Machine Learning
Token Hidden Reward: Steering Exploration-Exploitation in GRPO Training
Wenlong Deng, Yi Ren, Danica J. Sutherland, Christos Thrampoulidis, Xiaoxiao Li
2nd AI for Math Workshop
Best Paper Award