NeurIPS 2026
Drive vs. Decay has been accepted at NeurIPS 2026 (Main Track). The paper studies JEPA training dynamics and representation collapse. The main conference will take place in Sydney on December 6–12, 2026.
Flow Matching for Credit Card Fraud Detection
Evaluate one-step flow matching for credit card fraud detection, comparing classification quality, training cost and inference speed with gradient boosted trees.
Uncertainty quantification for large language models
This MMF project studies uncertainty estimates for transformer models, initially on tabular classification, with potential applications to the reliability of trading predictions.