Meta Fine-Tuning Neural Language Models for Multi-Domain Text Mining
This addresses the challenge of correlated learning across domains in NLP, offering a method to enhance fine-tuning for multi-domain text mining, though it appears incremental as it builds on existing fine-tuning approaches.
The paper tackles the problem of fine-tuning pre-trained language models for multi-domain NLP tasks by proposing Meta Fine-Tuning (MFT), which learns from typical instances across domains to improve transferability and generalization, resulting in better performance and usefulness for few-shot learning as confirmed by experiments on BERT.
Pre-trained neural language models bring significant improvement for various NLP tasks, by fine-tuning the models on task-specific training sets. During fine-tuning, the parameters are initialized from pre-trained models directly, which ignores how the learning process of similar NLP tasks in different domains is correlated and mutually reinforced. In this paper, we propose an effective learning procedure named Meta Fine-Tuning (MFT), served as a meta-learner to solve a group of similar NLP tasks for neural language models. Instead of simply multi-task training over all the datasets, MFT only learns from typical instances of various domains to acquire highly transferable knowledge. It further encourages the language model to encode domain-invariant representations by optimizing a series of novel domain corruption loss functions. After MFT, the model can be fine-tuned for each domain with better parameter initializations and higher generalization ability. We implement MFT upon BERT to solve several multi-domain text mining tasks. Experimental results confirm the effectiveness of MFT and its usefulness for few-shot learning.