Improving BERT Fine-Tuning via Self-Ensemble and Self-Distillation
This addresses the under-explored fine-tuning strategy for BERT, offering a method to enhance adaptation for NLP practitioners, though it appears incremental as it builds on existing fine-tuning approaches.
The paper tackles the problem of fine-tuning BERT for NLP tasks by proposing self-ensemble and self-distillation mechanisms, resulting in significant improvements on text classification and natural language inference tasks without external data.
Fine-tuning pre-trained language models like BERT has become an effective way in NLP and yields state-of-the-art results on many downstream tasks. Recent studies on adapting BERT to new tasks mainly focus on modifying the model structure, re-designing the pre-train tasks, and leveraging external data and knowledge. The fine-tuning strategy itself has yet to be fully explored. In this paper, we improve the fine-tuning of BERT with two effective mechanisms: self-ensemble and self-distillation. The experiments on text classification and natural language inference tasks show our proposed methods can significantly improve the adaption of BERT without any external data or knowledge.