mlsanityData to 7 Oct 2026

NeuroCounterfactuals: Beyond Minimal-Edit Counterfactuals for Richer Data Augmentation

arXiv 2210.12365 ↗First posted 22 Oct 2022cs.CLcited by 23

Not on the map. Why?

Abstract (arXiv)

While counterfactual data augmentation offers a promising step towards robust generalization in natural language processing, producing a set of counterfactuals that offer valuable inductive bias for models remains a challenge. Most existing approaches for producing counterfactuals, manual or automated, rely on small perturbations via minimal edits, resulting in simplistic changes. We introduce NeuroCounterfactuals, designed as loose counterfactuals, allowing for larger edits which result in naturalistic generations containing linguistic diversity, while still bearing similarity to the original document. Our novel generative approach bridges the benefits of constrained decoding, with those of language model adaptation for sentiment steering. Training data augmentation with our generations results in both in-domain and out-of-domain improvements for sentiment classification, outperforming even manually curated counterfactuals, under select settings. We further present detailed analyses to show the advantages of NeuroCounterfactuals over approaches involving simple, minimal edits.

Key facts. Cited by 23 papers in the map (data through 7 Oct 2026). Builds on 35 of its 59 references that are in the map. First posted on arXiv 22 Oct 2022.

Builds on

35 of 59 references.

References · 35

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2020Shortcut Learning in Deep Neural Networkscited by 1,928
  3. 2020Don't Stop Pretraining: Adapt Language Models to Domains and Taskscited by 1,548
  4. 2019Do ImageNet Classifiers Generalize to ImageNet?cited by 1,507
  5. 2020WILDS: A Benchmark of in-the-Wild Distribution Shiftscited by 1,366
  6. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  7. 2015A Diversity-Promoting Objective Function for Neural Conversation Modelscited by 1,068
  8. 2016ConceptNet 5.5: An Open Multilingual Graph of General Knowledgecited by 1,065
  9. 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 979
  10. 2019Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inferencecited by 864
  11. 2018Annotation Artifacts in Natural Language Inference Datacited by 656
  12. 2018Gender Bias in Coreference Resolutioncited by 438
23 more
  1. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 434
  2. 2019Learning the Difference that Makes a Difference with Counterfactually-Augmented Datacited by 428
  3. 2019MoverScore: Text Generation Evaluating with Contextualized Embeddings and Earth Mover Distancecited by 415
  4. 2021Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Modelscited by 219
  5. 2019SemEval-2017 Task 4: Sentiment Analysis in Twittercited by 209
  6. 2022WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creationcited by 200
  7. 2020Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Biascited by 183
  8. 2021Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyondcited by 177
  9. 2019Good-Enough Compositional Data Augmentationcited by 158
  10. 2020NeuroLogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraintscited by 146
  11. 2018Breaking NLI Systems with Sentences that Require Simple Lexical Inferencescited by 134
  12. 2020Generate Your Counterfactuals: Towards Controlled Counterfactual Generation for Textcited by 75
  13. 2020Robustness to Spurious Correlations in Text Classification via Automatically Generated Counterfactualscited by 67
  14. 2021Tailor: Generating and Perturbing Text with Semantic Controlscited by 67
  15. 2020Explaining The Efficacy of Counterfactually Augmented Datacited by 64
  16. 2020Robustness to Spurious Correlations via Human Annotationscited by 62
  17. 2022A Causal Lens for Controllable Text Generationcited by 45
  18. 2021An Investigation of the (In)effectiveness of Counterfactually Augmented Datacited by 42
  19. 2021Exploring the Efficacy of Automatically Generated Counterfactuals for Sentiment Analysiscited by 30
  20. 2021Retrieval-guided Counterfactual Generation for QAcited by 28
  21. 2020Counterfactually-Augmented SNLI Training Data Does Not Yield Better Generalization Than Unaugmented Datacited by 27
  22. 2022Informativeness and Invariance: Two Perspectives on Spurious Correlations in Natural Languagecited by 21
  23. 2021How Does Counterfactually Augmented Data Impact Models for Social Computing Constructs?cited by 19

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 15; 2024: 4; 2025: 2; 2026: 1.

23 papers.

Sort:

By area:

  1. 2022GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspectivecited by 72
  2. 2023SocialCounterfactuals: Probing and Mitigating Intersectional Social Biases in Vision-Language Models with Counterfactual Examplescited by 53
  3. 2023COCO-Counterfactuals: Automatically Constructed Counterfactual Examples for Image-Text Pairscited by 39
  4. 2023Prompting Large Language Models for Counterfactual Generation: An Empirical Studycited by 36
  5. 2023People Make Better Edits: Measuring the Efficacy of LLM-Generated Counterfactually Augmented Data for Harmful Language Detectioncited by 25
  6. 2024On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMscited by 24
  7. 2024A Survey on Natural Language Counterfactual Generationcited by 21
  8. 2024SK-VQA: Synthetic Knowledge Generation at Scale for Training Context-Augmented Multimodal LLMscited by 12
  9. 2023NeuroPrompts: An Adaptive Framework to Optimize Prompts for Text-to-Image Generationcited by 10
  10. 2023EXPLAIN, EDIT, GENERATE: Rationale-Sensitive Counterfactual Data Augmentation for Multi-hop Fact Verificationcited by 9
  11. 2023NeuroComparatives: Neuro-Symbolic Distillation of Comparative Knowledgecited by 8
  12. 2023SCENE: Self-Labeled Counterfactuals for Extrapolating to Negative Examplescited by 8
11 more
  1. 2023Controlling Learned Effects to Reduce Spurious Correlations in Text Classifierscited by 5
  2. 2023Navigating the Structured What-If Spaces: Counterfactual Generation via Structured Diffusioncited by 5
  3. 2025p-less Sampling: A Robust Hyperparameter-Free Approach for LLM Decodingcited by 4
  4. 2024Multi-Aspect Controllable Text Generation with Disentangled Counterfactual Augmentationcited by 3
  5. 2023Zero-Shot Cross-Lingual Sentiment Classification under Distribution Shift: an Exploratory Studycited by 2
  6. 2025Learning from Reasoning Failures via Synthetic Data Generationcited by 2
  7. 2023Counterfactual Augmentation for Multimodal Learning Under Presentation Biascited by 1
  8. 2023Out-of-Distribution Generalization in Text Classification: Past, Present, and Futurecited by 1
  9. 2023Investigating Masking-based Data Generation in Language Modelscited by 0
  10. 2023Targeted Image Data Augmentation Increases Basic Skills Captioning Robustnesscited by 0
  11. 2026CoCoGEC: Counterfactual Generation for Robust Grammatical Error Correctioncited by 0