Improving In-Context Few-Shot Learning via Self-Supervised Training
This work addresses the challenge of improving few-shot learning performance for NLP tasks, though it appears incremental as it builds on existing self-supervised pretraining methods.
The paper tackles the problem of adapting self-supervised pretraining for in-context few-shot learning by introducing an intermediate training stage with self-supervised objectives, resulting in models that outperform strong baselines on two benchmarks.
Self-supervised pretraining has made few-shot learning possible for many NLP tasks. But the pretraining objectives are not typically adapted specifically for in-context few-shot learning. In this paper, we propose to use self-supervision in an intermediate training stage between pretraining and downstream few-shot usage with the goal to teach the model to perform in-context few shot learning. We propose and evaluate four self-supervised objectives on two benchmarks. We find that the intermediate self-supervision stage produces models that outperform strong baselines. Ablation study shows that several factors affect the downstream performance, such as the amount of training data and the diversity of the self-supervised objectives. Human-annotated cross-task supervision and self-supervision are complementary. Qualitative analysis suggests that the self-supervised-trained models are better at following task requirements.