# Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models

arXiv: https://arxiv.org/abs/2101.00288 · HTML: https://mlsanity.com/paper/2101.00288/

## Key facts

- Cited by 222 papers in the map (data through 14 Oct 2026).
- Builds on 39 of its 66 references that are in the map.
- First posted on arXiv 1 Jan 2021.

## Builds on (most cited first, 30 of 39)

- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [SQuAD: 100,000+ Questions for Machine Comprehension of Text](https://mlsanity.com/paper/1606.05250/) (2016)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [Explanation in Artificial Intelligence: Insights from the Social Sciences](https://mlsanity.com/paper/1706.07269/) (2017)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Plug and Play Language Models: A Simple Approach to Controlled Text Generation](https://mlsanity.com/paper/1912.02164/) (2019)
- [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://mlsanity.com/paper/1909.05858/) (2019)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [Attention is not not Explanation](https://mlsanity.com/paper/1908.04626/) (2019)
- [Texygen: A Benchmarking Platform for Text Generation Models](https://mlsanity.com/paper/1802.01886/) (2018)
- [Learning the Difference that Makes a Difference with Counterfactually-Augmented Data](https://mlsanity.com/paper/1909.12434/) (2019)
- [PAWS: Paraphrase Adversaries from Word Scrambling](https://mlsanity.com/paper/1904.01130/) (2019)
- [Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance](https://mlsanity.com/paper/2006.14779/) (2020)
- [Adversarial Example Generation with Syntactically Controlled Paraphrase Networks](https://mlsanity.com/paper/1804.06059/) (2018)
- [Stress Test Evaluation for Natural Language Inference](https://mlsanity.com/paper/1806.00692/) (2018)
- [Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?](https://mlsanity.com/paper/2005.01831/) (2020)
- [Counterfactual Fairness in Text Classification through Robustness](https://mlsanity.com/paper/1809.10610/) (2018)
- [Contextualized Perturbation for Textual Adversarial Attack](https://mlsanity.com/paper/2009.07502/) (2020)
- [Breaking NLI Systems with Sentences that Require Simple Lexical Inferences](https://mlsanity.com/paper/1805.02266/) (2018)
- [Enabling Language Models to Fill in the Blanks](https://mlsanity.com/paper/2005.05339/) (2020)
- [Learning to Deceive with Attention-Based Explanations](https://mlsanity.com/paper/1909.07913/) (2019)
- [Explaining NLP Models via Minimal Contrastive Editing (MiCE)](https://mlsanity.com/paper/2012.13985/) (2020)
- [Counterfactual Story Reasoning and Generation](https://mlsanity.com/paper/1909.04076/) (2019)
- [Yelp Dataset Challenge: Review Rating Prediction](https://mlsanity.com/paper/1605.05362/) (2016)

## Built on by (most cited first, 30 of 222)

- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Discovering Language Model Behaviors with Model-Written Evaluations](https://mlsanity.com/paper/2212.09251/) (2022)
- [Recent Advances in Natural Language Processing via Large Pre-Trained Language Models: A Survey](https://mlsanity.com/paper/2111.01243/) (2021)
- [Explainability for Large Language Models: A Survey](https://mlsanity.com/paper/2309.01029/) (2023)
- [AI Chains: Transparent and Controllable Human-AI Interaction by Chaining Large Language Model Prompts](https://mlsanity.com/paper/2110.01691/) (2021)
- [Clever Hans or Neural Theory of Mind? Stress Testing Social Reasoning in Large Language Models](https://mlsanity.com/paper/2305.14763/) (2023)
- [Post-hoc Interpretability for Neural NLP: A Survey](https://mlsanity.com/paper/2108.04840/) (2021)
- [Rich Knowledge Sources Bring Complex Knowledge Conflicts: Recalibrating Models to Reflect Conflicting Evidence](https://mlsanity.com/paper/2210.13701/) (2022)
- [Benchmarking and Survey of Explanation Methods for Black Box Models](https://mlsanity.com/paper/2102.13076/) (2021)
- [Towards Faithful Model Explanation in NLP: A Survey](https://mlsanity.com/paper/2209.11326/) (2022)
- [Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing](https://mlsanity.com/paper/2102.12060/) (2021)
- [Measure and Improve Robustness in NLP Models: A Survey](https://mlsanity.com/paper/2112.08313/) (2021)
- [Faithfulness Tests for Natural Language Explanations](https://mlsanity.com/paper/2305.18029/) (2023)
- [Are self-explanations from Large Language Models faithful?](https://mlsanity.com/paper/2401.07927/) (2024)
- [Revisiting Out-of-distribution Robustness in NLP: Benchmark, Analysis, and LLMs Evaluations](https://mlsanity.com/paper/2306.04618/) (2023)
- [Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations](https://mlsanity.com/paper/2307.08678/) (2023)
- [Counterfactual Invariance to Spurious Correlations: Why and How to Pass Stress Tests](https://mlsanity.com/paper/2106.00545/) (2021)
- [Contrastive Explanations for Model Interpretability](https://mlsanity.com/paper/2103.01378/) (2021)
- [EvalLM: Interactive Evaluation of Large Language Model Prompts on User-Defined Criteria](https://mlsanity.com/paper/2309.13633/) (2023)
- [NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation](https://mlsanity.com/paper/2112.02721/) (2021)
- [GLUE-X: Evaluating Natural Language Understanding Models from an Out-of-distribution Generalization Perspective](https://mlsanity.com/paper/2211.08073/) (2022)
- [Tailor: Generating and Perturbing Text with Semantic Controls](https://mlsanity.com/paper/2107.07150/) (2021)
- [Interactive Debugging and Steering of Multi-Agent AI Systems](https://mlsanity.com/paper/2503.02068/) (2025)
- [From Language Modeling to Instruction Following: Understanding the Behavior Shift in LLMs after Instruction Tuning](https://mlsanity.com/paper/2310.00492/) (2023)
- [Leveraging Explanations in Interactive Machine Learning: An Overview](https://mlsanity.com/paper/2207.14526/) (2022)
- [SocialCounterfactuals: Probing and Mitigating Intersectional Social Biases in Vision-Language Models with Counterfactual Examples](https://mlsanity.com/paper/2312.00825/) (2023)
- [Perturbation Augmentation for Fairer NLP](https://mlsanity.com/paper/2205.12586/) (2022)
- [TIBET: Identifying and Evaluating Biases in Text-to-Image Generative Models](https://mlsanity.com/paper/2312.01261/) (2023)
- [Towards LLM-guided Causal Explainability for Black-box Text Classifiers](https://mlsanity.com/paper/2309.13340/) (2023)
- [An Investigation of the (In)effectiveness of Counterfactually Augmented Data](https://mlsanity.com/paper/2107.00753/) (2021)

## Abstract (arXiv)

While counterfactual examples are useful for analysis and training of NLP models, current generation methods either rely on manual labor to create very few counterfactuals, or only instantiate limited types of perturbations such as paraphrases or word substitutions. We present Polyjuice, a general-purpose counterfactual generator that allows for control over perturbation types and locations, trained by finetuning GPT-2 on multiple datasets of paired sentences. We show that Polyjuice produces diverse sets of realistic counterfactuals, which in turn are useful in various distinct applications: improving training and evaluation on three different tasks (with around 70% less annotation effort than manual generation), augmenting state-of-the-art explanation techniques, and supporting systematic counterfactual error analysis by revealing behaviors easily missed by human experts.