Selective In-Context Data Augmentation for Intent Detection using Pointwise V-Information

TL;DR

This paper introduces a novel data augmentation method for intent detection that uses large pre-trained language models and pointwise V-information to generate and filter synthetic data, improving performance especially in few-shot scenarios.

Contribution

The paper proposes a new approach combining PLMs and PVI for intent data augmentation, with intent-aware filtering to enhance classifier training.

Findings

Achieves state-of-the-art results in few-shot intent detection.

Performs comparably to full-data models in full-shot settings.

Demonstrates effective synthetic data generation and filtering.

Abstract

This work focuses on in-context data augmentation for intent detection. Having found that augmentation via in-context prompting of large pre-trained language models (PLMs) alone does not improve performance, we introduce a novel approach based on PLMs and pointwise V-information (PVI), a metric that can measure the usefulness of a datapoint for training a model. Our method first fine-tunes a PLM on a small seed of training data and then synthesizes new datapoints - utterances that correspond to given intents. It then employs intent-aware filtering, based on PVI, to remove datapoints that are not helpful to the downstream intent classifier. Our method is thus able to leverage the expressive power of large language models to produce diverse training data. Empirical results demonstrate that our method can produce synthetic training data that achieve state-of-the-art performance on three challenging intent detection datasets under few-shot settings (1.28% absolute improvement in 5-shot and 1.18% absolute in 10-shot, on average) and perform on par with the state-of-the-art in full-shot settings (within 0.01% absolute, on average).