Optimal and efficient text counterfactuals using Graph Neural Networks
Not on the map. Why?
Abstract (arXiv)
As NLP models become increasingly integral to decision-making processes, the need for explainability and interpretability has become paramount. In this work, we propose a framework that achieves the aforementioned by generating semantically edited inputs, known as counterfactual interventions, which change the model prediction, thus providing a form of counterfactual explanations for the model. We test our framework on two NLP tasks - binary sentiment classification and topic classification - and show that the generated edits are contrastive, fluent and minimal, while the whole process remains significantly faster that other state-of-the-art counterfactual editors.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 32 of its 62 references that are in the map. First posted on arXiv 4 Aug 2024.
Builds on
32 of 62 references.
References · 32
- 2017Attention Is All You Needcited by 59,932
- 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2019BERTScore: Evaluating Text Generation with BERTcited by 5,462
- 2013Intriguing properties of neural networkscited by 5,114
- 2019A Comprehensive Survey on Graph Neural Networkscited by 2,906
- 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,201
- 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 663
- 2017Adversarial Examples for Evaluating Reading Comprehension Systemscited by 568
- 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 488
- 2019Learning the Difference that Makes a Difference with Counterfactually-Augmented Datacited by 428
20 more
- 2020BAE: BERT-based Adversarial Examples for Text Classificationcited by 316
- 2018Adversarial Example Generation with Syntactically Controlled Paraphrase Networkscited by 282
- 2021Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Modelscited by 219
- 2020A Survey of the State of Explainable AI for Natural Language Processingcited by 206
- 2021Post-hoc Interpretability for Neural NLP: A Surveycited by 155
- 2020Contextualized Perturbation for Textual Adversarial Attackcited by 136
- 2020Explaining NLP Models via Minimal Contrastive Editing (MiCE)cited by 105
- 2022Interpreting Language Models with Contrastive Explanationscited by 91
- 2021Tailor: Generating and Perturbing Text with Semantic Controlscited by 67
- 2019On Evaluation of Adversarial Perturbations for Sequence-to-Sequence Modelscited by 56
- 2024GISTEmbed: Guided In-sample Selection of Training Negatives for Text Embedding Fine-tuningcited by 40
- 2024Multi-Task Contrastive Learning for 8192-Token Bilingual Text Embeddingscited by 20
- 2023CATfOOD: Counterfactual Augmented Training for Improving Out-of-Domain Performance and Calibrationcited by 16
- 2019Weight of Evidence as a Basis for Human-Oriented Explanationscited by 10
- 2021Detecting Textual Adversarial Examples through Randomized Substitution and Votecited by 8
- 2022GLAN: A Graph-based Linear Assignment Networkcited by 6
- 2023Counterfactuals of Counterfactuals: a back-translation-inspired approach to analyse counterfactual editorscited by 6
- 2022Towards explainable evaluation of language models on the semantic similarity of visual conceptscited by 5
- 2023BeamAttack: Generating High-quality Textual Adversarial Examples through Beam Search and Mixed Semantic Spacescited by 5
- 2022Identifying Human Strategies for Generating Word-Level Adversarial Examplescited by 2
Built on by
Citing papers per year: 2025: 1.
1 paper.