EMNLP 2023long findings0 citations

APP: Adaptive Prototypical Pseudo-Labeling for Few-shot OOD Detection

Pei Wang, Keqing He, Yutao Mou, Xiaoshuai Song, Yanan Wu, Jingang Wang, Yunsen Xian, Xunliang Cai

Abstract

Detecting out-of-domain (OOD) intents from user queries is essential for a task-oriented dialogue system. Previous OOD detection studies generally work on the assumption that plenty of labeled IND intents exist. In this paper, we focus on a more practical few-shot OOD setting where there are only a few labeled IND data and massive unlabeled mixed data that may belong to IND or OOD. The new scenario carries two key challenges: learning discriminative representations using limited IND data and leveraging unlabeled mixed data. Therefore, we propose an adaptive prototypical pseudo-labeling(APP) method for few-shot OOD detection, including a prototypical OOD detection framework (ProtoOOD) to facilitate low-resourceOOD detection using limited IND data, and an adaptive pseudo-labeling method to produce high-quality pseudo OOD and IND labels. Extensive experiments and analysis demonstrate the effectiveness of our method for few-shot OOD detection.

OODIntent DetectionFew-shotPrototype
BibTeX
@inproceedings{
wang2023app,
title={{APP}: Adaptive Prototypical Pseudo-Labeling for Few-shot {OOD} Detection},
author={Pei Wang and Keqing He and Yutao Mou and Xiaoshuai Song and Yanan Wu and Jingang Wang and Yunsen Xian and Xunliang Cai and Weiran Xu},
booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
year={2023},
url={https://openreview.net/forum?id=mrD5HN7ZNR}
}
APP: Adaptive Prototypical Pseudo-Labeling for Few-shot OOD Detection · EMNLP 2023