Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inference

TL;DR

This paper introduces a discriminative nearest neighbor approach with NLI transfer for few-shot intent detection, significantly improving accuracy and stability in out-of-scope detection while maintaining constant inference time.

Contribution

It presents a novel NLI transfer method combined with deep self-attention for few-shot intent detection, outperforming existing classifiers and embedding methods.

Findings

Achieves higher accuracy in in-domain and OOS detection than baselines.

Enables 10-shot models to match 50-shot or full-shot performance.

Maintains constant inference time with faster embedding retrieval.

Abstract

Intent detection is one of the core components of goal-oriented dialog systems, and detecting out-of-scope (OOS) intents is also a practically important skill. Few-shot learning is attracting much attention to mitigate data scarcity, but OOS detection becomes even more challenging. In this paper, we present a simple yet effective approach, discriminative nearest neighbor classification with deep self-attention. Unlike softmax classifiers, we leverage BERT-style pairwise encoding to train a binary classifier that estimates the best matched training example for a user input. We propose to boost the discriminative ability by transferring a natural language inference (NLI) model. Our extensive experiments on a large-scale multi-domain intent detection task show that our method achieves more stable and accurate in-domain and OOS detection accuracy than RoBERTa-based classifiers and embedding-based nearest neighbor approaches. More notably, the NLI transfer enables our 10-shot model to perform competitively with 50-shot or even full-shot classifiers, while we can keep the inference time constant by leveraging a faster embedding retrieval model.