1

LLM judges are easier to talk down than up: a Bayesian item response analysis

What does a Bayes-filtered transformer believe? A predictive Monte Carlo approach

TabMGP: Martingale Posterior with TabPFN

Temperature Optimization for Bayesian Deep Learning

A Bayesian Model Selection Criterion for Selecting Pretraining Checkpoints

There are two arXiv versions with different titles, but the material is very similar; v2 matches the ICML camera-ready version.

The Local Learning Coefficient: A Singularity-Aware Complexity Measure

arXiv v1 appeared under a different title and author list, with a stronger emphasis on mathematical development. arXiv v2 matches the AISTATS camera-ready paper and integrates material from [arXiv:2402.03698](https://arxiv.org/abs/2402.03698).

Pathwise Gradient Variance Reduction with Control Variates in Variational Inference

Interventional Fairness on Partially Known Causal Graphs -- A Constrained Optimization Approach

Counterfactual Fairness with Partially Known Causal Graph

A Shooting Formulation of Deep Learning

Selected for oral presentation