Marika NiihoriWe are delighted to share that seven papers from the van der Schaar Lab have been accepted to the International Conference on Learning Representations (ICLR) 2026.
ICLR is one of the leading international conferences in machine learning, with a strong focus on representation learning, deep learning, and the theoretical and practical foundations of modern AI. Each year, it brings together researchers from across academia and industry as well as engineers and entrepreneurs to present advances that shape the future of the field.
This year from the lab, works span new ML advances on autoformalism (a new ML research area which we are building), causality, diffusion models, deep learning models and LLMs – reflecting both the breadth and depth of our lab’s research. These are listed below:
- MATHMO: Automated Mathematical Modeling Through Adaptive Search
Tennison Liu, Mihaela van der Schaar - Operator Theory-Driven Autoformulation of MDPs for Control of Queueing Systems
Victor Baillet, Yuanzhang Xiao, Nicolás Astorga, Mihaela van der Schaar - Hyperparameter Trajectory Inference with Conditional Lagrangian Optimal Transport
Harry Amad, Mihaela van der Schaar - Eliciting Numerical Predictive Distributions of LLMs Without Auto-Regression
Julianna Piskorz, Kasia Kobalczyk, Mihaela van der Schaar - Deep Hierarchical Learning with Nested Subspace Networks
Paulius Rauba, Mihaela van der Schaar - A Study of Posterior Stability in Time-Series Latent Diffusion
Yangming Li, Yixin Cheng, Mihaela van der Schaar - Overlap-weighted orthogonal meta-learner for treatment effect estimation over time
Konstantin Hess, Dennis Frauen, Mihaela van der Schaar, Stefan Feuerriegel
Congratulations to all co-authors and collaborators! We look forward to sharing more details on the highlights in the coming few weeks.
