mlsanityData to 7 Oct 2026

ProtoInfoMax: Prototypical Networks with Mutual Information Maximization for Out-of-Domain Detection

arXiv 2108.12229 ↗First posted 27 Aug 2021cs.CLcited by 4

Not on the map. Why?

Abstract (arXiv)

The ability to detect Out-of-Domain (OOD) inputs has been a critical requirement in many real-world NLP applications. For example, intent classification in dialogue systems. The reason is that the inclusion of unsupported OOD inputs may lead to catastrophic failure of systems. However, it remains an empirical question whether current methods can tackle such problems reliably in a realistic scenario where zero OOD training data is available. In this study, we propose ProtoInfoMax, a new architecture that extends Prototypical Networks to simultaneously process in-domain and OOD sentences via Mutual Information Maximization (InfoMax) objective. Experimental results show that our proposed method can substantially improve performance up to 20% for OOD detection in low resource settings of text classification. We also show that ProtoInfoMax is less prone to typical overconfidence errors of Neural Networks, leading to more reliable prediction results.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 32 of its 51 references that are in the map. First posted on arXiv 27 Aug 2021.

Builds on

32 of 51 references.

References · 32

  1. 2017On Calibration of Modern Neural Networkscited by 4,860
  2. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,888
  3. 2017Prototypical Networks for Few-shot Learningcited by 3,001
  4. 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,325
  5. 2018Learning deep representations by mutual information estimation and maximizationcited by 1,265
  6. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,068
  7. 2018Deep Anomaly Detection with Outlier Exposurecited by 939
  8. 2018MINE: Mutual Information Neural Estimationcited by 680
  9. 2020Efficient Intent Detection with Dual Sentence Encoderscited by 419
  10. 2019An Evaluation Dataset for Intent Classification and Out-of-Scope Predictioncited by 378
  11. 2019On Mutual Information Maximization for Representation Learningcited by 315
  12. 2018Recent Advances in Open Set Recognition: A Surveycited by 309
20 more
  1. 2018FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluationcited by 254
  2. 2019OCGAN: One-class Novelty Detection Using GANs with Constrained Latent Representationscited by 199
  3. 2018Adversarially Learned Anomaly Detectioncited by 115
  4. 2017DOC: Deep Open Classification of Text Documentscited by 109
  5. 2019FewRel 2.0: Towards More Challenging Few-Shot Relation Classificationcited by 94
  6. 2019A Mutual Information Maximization Perspective of Language Representation Learningcited by 85
  7. 2017Learning how to Active Learn: A Deep Reinforcement Learning Approachcited by 79
  8. 2013One-Class Classification: Taxonomy of Study and Review of Techniquescited by 68
  9. 2019Out-of-domain Detection for Natural Language Understanding in Dialog Systemscited by 62
  10. 2018Diverse Few-Shot Text Classification with Multiple Metricscited by 61
  11. 2019Few-shot Text Classification with Distributional Signaturescited by 58
  12. 2020Discriminative Nearest Neighbor Few-Shot Intent Detection by Transferring Natural Language Inferencecited by 47
  13. 2019Out-of-Domain Detection for Low-Resource Text Classification Taskscited by 36
  14. 2018Learning Speaker Representations with Mutual Informationcited by 26
  15. 2020Intent Mining from past conversations for conversational agentcited by 25
  16. 2019Distance-Based Learning from Errors for Confidence Calibrationcited by 23
  17. 2018Neural Sentence Embedding using Only In-domain Sentences for Out-of-domain Sentence Detection in Dialog Systemscited by 18
  18. 2018Joint Learning of Domain Classification and Out-of-Domain Detection with Dynamic Class Weighting for Satisficing False Acceptance Ratescited by 17
  19. 2019QAInfomax: Learning Robust Question Answering System by Mutual Information Maximizationcited by 9
  20. 2021A Neural Few-Shot Text Classification Reality Checkcited by 6

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2.

4 papers.

  1. 2023Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detectioncited by 29
  2. 2024Domain Adaptation in Intent Classification Systems: A Reviewcited by 4
  3. 2024ROGPL: Robust Open-Set Graph Learning via Region-Based Prototype Learningcited by 2
  4. 2023KeyGen2Vec: Learning Document Embedding via Multi-label Keyword Generation in Question-Answeringcited by 0