Example-Driven Intent Prediction with Observers

TL;DR

This paper introduces observer tokens and example-driven training to enhance intent classification in dialog systems, achieving state-of-the-art results and strong transferability in few-shot and cross-dataset scenarios.

Contribution

It proposes novel observer tokens and an example-driven training approach to improve generalization and transferability of intent classification models.

Findings

Achieved state-of-the-art results on three intent datasets.

Improved few-shot learning performance with the proposed methods.

Demonstrated effective transfer to new intents and datasets without retraining.

Abstract

A key challenge of dialog systems research is to effectively and efficiently adapt to new domains. A scalable paradigm for adaptation necessitates the development of generalizable models that perform well in few-shot settings. In this paper, we focus on the intent classification problem which aims to identify user intents given utterances addressed to the dialog system. We propose two approaches for improving the generalizability of utterance classification models: (1) observers and (2) example-driven training. Prior work has shown that BERT-like models tend to attribute a significant amount of attention to the [CLS] token, which we hypothesize results in diluted representations. Observers are tokens that are not attended to, and are an alternative to the [CLS] token as a semantic representation of utterances. Example-driven training learns to classify utterances by comparing to examples, thereby using the underlying encoder as a sentence similarity model. These methods are complementary; improving the representation through observers allows the example-driven model to better measure sentence similarities. When combined, the proposed methods attain state-of-the-art results on three intent prediction datasets (\textsc{banking77}, \textsc{clinc150}, \textsc{hwu64}) in both the full data and few-shot (10 examples per intent) settings. Furthermore, we demonstrate that the proposed approach can transfer to new intents and across datasets without any additional training.