mlsanityData to 14 Oct 2026

A Neural Few-Shot Text Classification Reality Check

arXiv 2101.12073 ↗First posted 28 Jan 2021cs.CLcited by 6

Not on the map. Why?

Abstract (arXiv)

Modern classification models tend to struggle when the amount of annotated data is scarce. To overcome this issue, several neural few-shot classification models have emerged, yielding significant progress over time, both in Computer Vision and Natural Language Processing. In the latter, such models used to rely on fixed word embeddings before the advent of transformers. Additionally, some models used in Computer Vision are yet to be tested in NLP applications. In this paper, we compare all these models, first adapting those made in the field of image processing to NLP, and second providing them access to transformers. We then test these models equipped with the same transformer-based encoder on the intent detection task, known for having a large number of classes. Our results reveal that while methods perform almost equally on the ARSC dataset, this is not the case for the Intent Detection task, where the most recent and supposedly best competitors perform worse than older and simpler ones (while all are given access to transformers). We also show that a simple baseline is surprisingly strong. All the new developed models, as well as the evaluation framework, are made publicly available.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Builds on 15 of its 21 references that are in the map. First posted on arXiv 28 Jan 2021.

Builds on

15 of 21 references.

References · 15

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2017Prototypical Networks for Few-shot Learningcited by 3,015
  3. 2016Matching Networks for One Shot Learningcited by 2,255
  4. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,143
  5. 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,060
  6. 2019BERT Rediscovers the Classical NLP Pipelinecited by 1,059
  7. 2015Improving Neural Machine Translation Models with Monolingual Datacited by 788
  8. 2017Dynamic Routing Between Capsulescited by 689
  9. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  10. 2019How to Fine-Tune BERT for Text Classification?cited by 411
  11. 2019An Evaluation Dataset for Intent Classification and Out-of-Scope Predictioncited by 386
  12. 2019Benchmarking Natural Language Understanding Services for building Conversational Agentscited by 166
3 more
  1. 2019Induction Networks for Few-Shot Text Classificationcited by 64
  2. 2018Diverse Few-Shot Text Classification with Multiple Metricscited by 61
  3. 2019Learning Only from Relevant Keywords and Unlabeled Documentscited by 3

Built on by

Cited by year

Citing papers per year: 2021: 2; 2022: 3; 2023: 1.

6 papers.

  1. 2021ProtAugment: Unsupervised diverse short-texts paraphrasing for intent detection meta-learningcited by 18
  2. 2022Improving Meta-learning for Low-resource Text Classification and Generation via Memory Imitationcited by 8
  3. 2023Pre-training Intent-Aware Encoders for Zero- and Few-Shot Intent Classificationcited by 5
  4. 2021ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detectioncited by 4
  5. 2022Empowering Sentence Encoders with Prompting and Label Retrieval for Zero-shot Text Classificationcited by 2
  6. 2022ProtSi: Prototypical Siamese Network with Data Augmentation for Few-Shot Subjective Answer Evaluationcited by 0