Showing cs.LGShow all
2 papers · 1 filter
cs.LG2026
Counterfactual Test-Time Augmentation for Shortcut Removal in Language Models
Katie Matton, Rosalind W. Picard, John Guttag +1
Large language models (LLMs) can exploit shortcuts, i.e., correlations present in training data that may not hold at test time. This can lead to poor predictions when correlations…
cs.LG2026
Improving Domain Generalization in Contrastive Learning using Adaptive Temperature Control
Robert Lewis, Katie Matton, Rosalind W. Picard +1
Self-supervised pre-training with contrastive learning is a powerful method for learning from sparsely labeled data. However, performance can drop considerably when there is a shif…