Out-of-Scope Intent Detection with Self-Supervision and Discriminative Training

TL;DR

This paper introduces a simple, end-to-end method for out-of-scope intent detection in dialogue systems that uses synthetic pseudo outliers for training, eliminating the need for distribution assumptions or post-processing.

Contribution

It proposes a novel training approach that simulates test scenarios with pseudo outliers, improving out-of-scope detection without complex procedures.

Findings

Significant performance improvements over state-of-the-art methods.

Effective generalization to real out-of-scope utterances.

No need for distribution assumptions or threshold tuning.

Abstract

Out-of-scope intent detection is of practical importance in task-oriented dialogue systems. Since the distribution of outlier utterances is arbitrary and unknown in the training stage, existing methods commonly rely on strong assumptions on data distribution such as mixture of Gaussians to make inference, resulting in either complex multi-step training procedures or hand-crafted rules such as confidence threshold selection for outlier detection. In this paper, we propose a simple yet effective method to train an out-of-scope intent classifier in a fully end-to-end manner by simulating the test scenario in training, which requires no assumption on data distribution and no additional post-processing or threshold setting. Specifically, we construct a set of pseudo outliers in the training stage, by generating synthetic outliers using inliner features via self-supervision and sampling out-of-scope sentences from easily available open-domain datasets. The pseudo outliers are used to train a discriminative classifier that can be directly applied to and generalize well on the test task. We evaluate our method extensively on four benchmark dialogue datasets and observe significant improvements over state-of-the-art approaches. Our code has been released at https://github.com/liam0949/DCLOOS.