
Enhancing Language Model Generalization: Bridging the Gap Between In-Context Learning and Fine-Tuning
Language models (LMs) have great capabilities as in-context learners when pretrained on vast internet text corpora, allowing them to generalize effectively from just a few task examples. However, fine-tuning these models for downstream tasks presents […]