1

Improving Data-Constrained Scaling with Hyperparameter Transfer and Ensembling
As growing compute budgets outpace the growth of high-quality training data, large language models must be trained with both compute …
Ethan Baron, Shmuel Berman, Anirudh Satheesh, Akshay Vegesna, Samip Dahal, Andrew Gordon Wilson
arXiv GitHub
Deep Generative Models for Phylogenetic Inference with Complex Evolutionary Processes
Phylogenetic inference plays a central role in understanding evolutionary relationships, with applications ranging from tracking …
Ethan Baron*, Alan Nawzad Amin*, Andrew Gordon Wilson
Spotlight @ NeurIPS Oral @ MLCB Spotlight @ ICML GenBio Paper @ ICML SPIGM Paper @ ICML GFM arXiv GitHub
Boulder2Vec: Modeling Climber Performances in Professional Bouldering Competitions
Using data from professional bouldering competitions from 2008 to 2022, we train a logistic regression to predict climber results and …
Ethan Baron, Victor Hau, Zeke Weng
Best Student Paper @ CMSAC Talk @ CMSAC arXiv GitHub
A Diffusion Model to Shrink Proteins While Maintaining Their Function
Many proteins useful in modern medicine or bioengineering are challenging to make in the lab, fuse with other proteins in cells, or …
Ethan Baron*, Alan Nawzad Amin*, Ruben Weitzman, Debora Susan Marks, Andrew Gordon Wilson
Paper @ ICLR Video @ ICLR Best Paper @ MoML Best Paper @ ICML EXAIT Spotlight @ ICML GenBio Paper @ MLCB arXiv GitHub
Efficiently Generating Correlated Sample Paths from Multi-Step Time Series Foundation Models
Many time series applications require access to multi-step forecast trajectories in the form of sample paths. Recently, time series …
Ethan Baron, Boris Oreshkin, Ruijun Ma, Hanyu Zhang, Kari Torkkola, Michael W. Mahoney, Andrew Gordon Wilson, Tatiana Konstantinova
Top 4 Paper @ NeurIPS BERT2S arXiv
Bike2Vec: Vector Embedding Representations of Road Cycling Riders and Races
Vector embeddings have been successfully applied in several domains to obtain effective representations of non-numeric data which can …
Ethan Baron, Bram Janssens, Matthias Bogaert
Paper @ MathSport arXiv GitHub
Multi-Aspect Reviewed-Item Retrieval via LLM Query Decomposition and Aspect Fusion
While user-generated product reviews contain a large amount of information, their utility in addressing natural language product …
Anton Korikov, George Saad, Ethan Baron, Mustafa Khan, Manav Shah and Scott Sanner
Oral @ SIGIR IR-RAG Paper @ SIGIR IR-RAG arXiv
Modelling GTHA Post-Secondary School Location Choice
We modelled school location choices of post-secondary students in the Greater Toronto and Hamilton Area using data from two student …
Ethan Baron, Gonzalo Martinez Santos, Eric J. Miller
Paper @ TAC GitHub