Fine-tuning Pre-trained Language Models for Few-shot Intent Detection: Supervised Pre-training and Isotropization
This work addresses the problem of limited labeled data for intent classification in dialogue systems, offering an incremental improvement over existing fine-tuning methods.
The paper tackles the challenge of few-shot intent detection in task-oriented dialogue systems by addressing the anisotropic feature space issue in supervised pre-training of language models, proposing isotropization regularizers that improve performance, with experimental results showing effectiveness.
It is challenging to train a good intent classifier for a task-oriented dialogue system with only a few annotations. Recent studies have shown that fine-tuning pre-trained language models with a small amount of labeled utterances from public benchmarks in a supervised manner is extremely helpful. However, we find that supervised pre-training yields an anisotropic feature space, which may suppress the expressive power of the semantic representations. Inspired by recent research in isotropization, we propose to improve supervised pre-training by regularizing the feature space towards isotropy. We propose two regularizers based on contrastive learning and correlation matrix respectively, and demonstrate their effectiveness through extensive experiments. Our main finding is that it is promising to regularize supervised pre-training with isotropization to further improve the performance of few-shot intent detection. The source code can be found at https://github.com/fanolabs/isoIntentBert-main.