Showing 2026Show all
2 papers · 1 filter
cs.LG2026
Train for Many, Update with One: Improving Multi-Sample Reasoning via Fenchel-Bregman Posterior Learning
Sungjun Lim, Adrian Robert Minut, Nico Carlsen Daheim +3
Multi-sample reasoning measures candidate coverage, the probability that at least one of candidates is correct. However, minimizing the average loss of individual attempts does…
cs.LG2026
Geometry-Adaptive Explainer for Faithful Dictionary-Based Interpretability under Distribution Shift
Sungjun Lim, Heedong Kim, Andrew Lee +1
Mechanistic interpretability aims to explain a model's behavior by identifying causally responsible internal structures. Dictionary-based explainers such as sparse autoencoders and…