mlsanityData to 7 Oct 2026

ProtAugment: Unsupervised diverse short-texts paraphrasing for intent detection meta-learning

arXiv 2105.12995 ↗First posted 27 May 2021cs.CL cs.AI cs.LGcited by 18

Not on the map. Why?

Abstract (arXiv)

Recent research considers few-shot intent detection as a meta-learning problem: the model is learning to learn from a consecutive set of small tasks named episodes. In this work, we propose ProtAugment, a meta-learning algorithm for short texts classification (the intent detection task). ProtAugment is a novel extension of Prototypical Networks, that limits overfitting on the bias introduced by the few-shots classification objective at each episode. It relies on diverse paraphrasing: a conditional language model is first fine-tuned for paraphrasing, and diversity is later introduced at the decoding stage at each meta-learning episode. The diverse paraphrasing is unsupervised as it is applied to unlabelled data, and then fueled to the Prototypical Network training objective as a consistency loss. ProtAugment is the state-of-the-art method for intent detection meta-learning, at no extra labeling efforts and without the need to fine-tune a conditional language model on a given application domain.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 30 of its 44 references that are in the map. First posted on arXiv 27 May 2021.

Builds on

30 of 44 references.

References · 30

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  3. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  4. 2016Matching Networks for One Shot Learningcited by 2,246
  5. 2017Learning to Compare: Relation Network for Few-Shot Learningcited by 1,139
  6. 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,057
  7. 2019A Closer Look at Few-shot Classificationcited by 679
  8. 2018Meta-Learning for Semi-Supervised Few-Shot Classificationcited by 456
  9. 2020Efficient Intent Detection with Dual Sentence Encoderscited by 419
  10. 2019PAWS: Paraphrase Adversaries from Word Scramblingcited by 395
  11. 2019An Evaluation Dataset for Intent Classification and Out-of-Scope Predictioncited by 378
  12. 2018Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfacescited by 376
18 more
  1. 2019PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identificationcited by 262
  2. 2019Probing Neural Network Comprehension of Natural Language Argumentscited by 212
  3. 2019Benchmarking Natural Language Understanding Services for building Conversational Agentscited by 166
  4. 2021Free Lunch for Few-shot Learning: Distribution Calibrationcited by 165
  5. 2018Meta-Learning for Low-Resource Neural Machine Translationcited by 105
  6. 2020DialoGLUE: A Natural Language Understanding Benchmark for Task-Oriented Dialoguecited by 89
  7. 2019Comparison of Diverse Decoding Methods from Conditional Language Modelscited by 78
  8. 2019Induction Networks for Few-Shot Text Classificationcited by 64
  9. 2020Neural Syntactic Preordering for Controlled Paraphrase Generationcited by 56
  10. 2019Natural Language Understanding with the Quora Question Pairs Datasetcited by 51
  11. 2019Investigating Meta-Learning Algorithms for Low-Resource Natural Language Understanding Taskscited by 48
  12. 2018Importance of Search and Evaluation Strategies in Neural Dialogue Modelingcited by 44
  13. 2019Unsupervised Paraphrasing by Simulated Annealingcited by 37
  14. 2018Natural Language to Structured Query Generation via Meta-Learningcited by 33
  15. 2016Noisy Parallel Approximate Decoding for Conditional Recurrent Language Modelcited by 18
  16. 2020schuBERT: Optimizing Elements of BERTcited by 17
  17. 2021A Neural Few-Shot Text Classification Reality Checkcited by 6
  18. 2018Refining Source Representations with Relation Networks for Neural Machine Translationcited by 1

Built on by

Cited by year

Citing papers per year: 2021: 1; 2022: 4; 2023: 5; 2024: 5; 2025: 3.

18 papers.

Sort:

By area:

  1. 2023Paraphrasing evades detectors of AI-generated text, but retrieval is an effective defensecited by 401
  2. 2021NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentationcited by 72
  3. 2022Why is Winoground Hard? Investigating Failures in Visuolinguistic Compositionalitycited by 44
  4. 2022Human-Machine Collaboration Approaches to Build a Dialogue Dataset for Hate Speech Counteringcited by 40
  5. 2024SemEval-2024 Shared Task 6: SHROOM, a Shared-task on Hallucinations and Related Observable Overgeneration Mistakescited by 24
  6. 2023ContrastNet: A Contrastive Learning Framework for Few-Shot Text Classificationcited by 17
  7. 2025A survey of diversity quantification in natural language processing: The why, what, where and howcited by 4
  8. 2023TART: Improved Few-shot Text Classification Using Task-Adaptive Reference Transformationcited by 3
  9. 2022The Devil is in the Details: On Models and Training Regimes for Few-Shot Intent Classificationcited by 2
  10. 2024Can Your Model Tell a Negation from an Implicature? Unravelling Challenges With Intent Encoderscited by 2
  11. 2024Prototypical Extreme Multi-label Classification with a Dynamic Margin Losscited by 2
  12. 2023Adversarial Word Dilution as Text Data Augmentation in Low-Resource Regimecited by 1
6 more
  1. 2023Data Augmentation for Emotion Detection in Small Imbalanced Text Datacited by 1
  2. 2024Large Language Models on Fine-grained Emotion Detection Dataset with Data Augmentation and Transfer Learningcited by 1
  3. 2025HausaNLP at SemEval-2025 Task 3: Towards a Fine-Grained Model-Aware Hallucination Detectioncited by 1
  4. 2025REFORMER: A ChatGPT-Driven Data Synthesis Framework Elevating Text-to-SQL Modelscited by 1
  5. 2022ProtSi: Prototypical Siamese Network with Data Augmentation for Few-Shot Subjective Answer Evaluationcited by 0
  6. 2024Exploring Description-Augmented Dataless Intent Classificationcited by 0