mlsanityData to 14 Oct 2026

TreeMix: Compositional Constituency-based Data Augmentation for Natural Language Understanding

arXiv 2205.06153 ↗First posted 12 May 2022cs.CLcited by 15

Not on the map. Why?

Abstract (arXiv)

Data augmentation is an effective approach to tackle over-fitting. Many previous works have proposed different data augmentations strategies for NLP, such as noise injection, word replacement, back-translation etc. Though effective, they missed one important characteristic of language--compositionality, meaning of a complex expression is built from its sub-parts. Motivated by this, we propose a compositional data augmentation approach for natural language understanding called TreeMix. Specifically, TreeMix leverages constituency parsing tree to decompose sentences into constituent sub-structures and the Mixup data augmentation technique to recombine them to generate new sentences. Compared with previous approaches, TreeMix introduces greater diversity to the samples generated and encourages models to learn compositionality of NLP data. Extensive experiments on text classification and SCAN demonstrate that TreeMix outperforms current state-of-the-art data augmentation methods.

Key facts. Cited by 15 papers in the map (data through 14 Oct 2026). Builds on 50 of its 61 references that are in the map. First posted on arXiv 12 May 2022.

Builds on

50 of 61 references.

References · 50

  1. 2017mixup: Beyond Empirical Risk Minimizationcited by 5,183
  2. 2018GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understandingcited by 4,899
  3. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,884
  4. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  5. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  6. 2019Unsupervised Data Augmentation for Consistency Trainingcited by 1,060
  7. 2017Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learningcited by 920
  8. 2019EDA: Easy Data Augmentation Techniques for Boosting Performance on Text Classification Taskscited by 807
  9. 2015Improving Neural Machine Translation Models with Monolingual Datacited by 788
  10. 2021A Survey of Data Augmentation Approaches for NLPcited by 409
  11. 2019FreeLB: Enhanced Adversarial Training for Natural Language Understandingcited by 264
  12. 2020Puzzle Mix: Exploiting Saliency and Local Statistics for Optimal Mixupcited by 261
38 more
  1. 2019SMART: Robust and Efficient Fine-Tuning for Pre-trained Natural Language Models through Principled Regularized Optimizationcited by 254
  2. 2018Contextual Augmentation: Data Augmentation by Words with Paradigmatic Relationscited by 201
  3. 2020MixText: Linguistically-Informed Interpolation of Hidden Space for Semi-Supervised Text Classificationcited by 187
  4. 2021GPT3Mix: Leveraging Large-scale Language Models for Text Augmentationcited by 187
  5. 2019Not Enough Data? Deep Learning to the Rescue!cited by 175
  6. 2020Data Augmentation using Pre-trained Transformer Modelscited by 168
  7. 2018QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehensioncited by 164
  8. 2019Good-Enough Compositional Data Augmentationcited by 158
  9. 2019Revisiting Self-Training for Neural Sequence Generationcited by 155
  10. 2020SaliencyMix: A Saliency Guided Data Augmentation Strategy for Better Regularizationcited by 149
  11. 2017Data Augmentation for Low-Resource Neural Machine Translationcited by 139
  12. 2021Co-Mixup: Saliency Guided Joint Mixup with Supermodular Diversitycited by 128
  13. 2019Augmenting Data with Mixup for Sentence Classification: An Empirical Studycited by 125
  14. 2016Data Recombination for Neural Semantic Parsingcited by 111
  15. 2021An Empirical Survey of Data Augmentation for Limited Data Learning in NLPcited by 101
  16. 2020SSMBA: Self-Supervised Manifold Based Data Augmentation for Improving Out-of-Domain Robustnesscited by 96
  17. 2020Compositional Generalization via Neural-Symbolic Stack Machinescited by 80
  18. 2017Data Noising as Smoothing in Neural Network Language Modelscited by 78
  19. 2020Span-based Semantic Parsing for Compositional Generalizationcited by 77
  20. 2020An Analysis of Simple Data Augmentation for Named Entity Recognitioncited by 76
  21. 2018SwitchOut: an Efficient Data Augmentation Algorithm for Neural Machine Translationcited by 72
  22. 2020A Simple but Tough-to-Beat Data Augmentation Approach for Natural Language Understanding and Generationcited by 72
  23. 2020Sequence-Level Mixed Sample Data Augmentationcited by 71
  24. 2020AdvAug: Robust Adversarial Augmentation for Neural Machine Translationcited by 67
  25. 2021AEDA: An Easier Data Augmentation Technique for Text Classificationcited by 67
  26. 2020Data Boost: Text Data Augmentation Through Reinforcement Learning Guided Conditional Generationcited by 47
  27. 2020GenAug: Data Augmentation for Finetuning Text Generatorscited by 41
  28. 2021SSMix: Saliency-Based Span Mixup for Text Classificationcited by 39
  29. 2020Data Manipulation: Towards Effective Instance Learning for Neural Dialogue Generation via Learning to Augment and Reweightcited by 34
  30. 2020Local Additivity Based Data Augmentation for Semi-supervised NERcited by 31
  31. 2017Unsupervised Domain Adaptation for Robust Speech Recognition via Variational Autoencoder-Based Data Augmentationcited by 27
  32. 2020Snippext: Semi-supervised Opinion Mining with Augmented Datacited by 26
  33. 2021Substructure Substitution: Structured Data Augmentation for NLPcited by 25
  34. 2019Controlled Text Generation for Data Augmentation in Intelligent Artificial Agentscited by 21
  35. 2020Tell Me How to Ask Again: Question Data Augmentation with Controllable Rewriting in Continuous Spacecited by 19
  36. 2020On the Role of Supervision in Unsupervised Constituency Parsingcited by 18
  37. 2018On Tree-Based Neural Sentence Modelingcited by 16
  38. 2018Unsupervised Adaptation with Interpretable Disentangled Representations for Distant Conversational Speech Recognitioncited by 7

Built on by

Cited by year

Citing papers per year: 2022: 6; 2023: 5; 2024: 3; 2025: 1.

15 papers.

Sort:

By area:

  1. 2023Increasing Diversity While Maintaining Accuracy: Text Data Generation with Large Language Models and Human Interventionscited by 125
  2. 2023DyVal: Dynamic Evaluation of Large Language Models for Reasoning Taskscited by 101
  3. 2022Contrastive Learning Reduces Hallucination in Conversationscited by 40
  4. 2024A Comprehensive Survey on Data Augmentationcited by 40
  5. 2022A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainabilitycited by 31
  6. 2024A Survey on Mixup Augmentations and Beyondcited by 20
  7. 2022PromptDA: Label-guided Data Augmentation for Prompt-based Few-shot Learnerscited by 7
  8. 2023Data Augmentation for Conversational AIcited by 3
  9. 2025LM-mixup: Text Data Augmentation via Language Model based Mixupcited by 3
  10. 2022Augmentation with Projection: Towards an Effective and Efficient Data Augmentation Paradigm for Distillationcited by 2
  11. 2023Understanding the Effect of Data Augmentation on Knowledge Distillationcited by 2
  12. 2023Steering Language Generation: Harnessing Contrastive Expert Guidance and Negative Prompting for Coherent and Diverse Synthetic Data Generationcited by 2
3 more
  1. 2022Syntax-driven Data Augmentation for Named Entity Recognitioncited by 0
  2. 2022Feature Weaken: Vicinal Data Augmentation for Classificationcited by 0
  3. 2024Evaluation Metrics for Text Data Augmentation in NLPcited by 0