Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
On the Unreasonable Effectiveness of Last-layer Retraining
John C. Hill, Tyler LaBonte, Xinchen Zhang +1
Last-layer retraining (LLR) methods -- wherein the last layer of a neural network is reinitialized and retrained on a held-out set following ERM training -- have garnered interest…
cs.LG2025
Improved and Oracle-Efficient Online -Multicalibration
Rohan Ghuge, Vidya Muthukumar, Sahil Singla
We study \emph{online multicalibration}, a framework for ensuring calibrated predictions across multiple groups in adversarial settings, across rounds. Although online calibrat…
cs.LG2024
The Group Robustness is in the Details: Revisiting Finetuning under Spurious Correlations
Tyler LaBonte, John C. Hill, Xinchen Zhang +2
Modern machine learning models are prone to over-reliance on spurious correlations, which can often lead to poor performance on minority groups. In this paper, we identify surprisi…