Faithfulness and the Notion of Adversarial Sensitivity in NLP Explanations
Not on the map. Why?
Abstract (arXiv)
Faithfulness is arguably the most critical metric to assess the reliability of explainable AI. In NLP, current methods for faithfulness evaluation are fraught with discrepancies and biases, often failing to capture the true reasoning of models. We introduce Adversarial Sensitivity as a novel approach to faithfulness evaluation, focusing on the explainer's response when the model is under adversarial attack. Our method accounts for the faithfulness of explainers by capturing sensitivity to adversarial input changes. This work addresses significant limitations in existing evaluation techniques, and furthermore, quantifies faithfulness from a crucial yet underexplored paradigm.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 63 of its 98 references that are in the map. First posted on arXiv 26 Sep 2024.
Builds on
63 of 98 references.
References · 63
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2014Explaining and Harnessing Adversarial Examplescited by 7,364
- 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
- 2013Intriguing properties of neural networkscited by 5,155
- 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
- 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
- 2017Axiomatic Attribution for Deep Networkscited by 3,384
- 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,385
- 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
- 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
- 2017Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networkscited by 1,328
- 2017Explanation in Artificial Intelligence: Insights from the Social Sciencescited by 1,206
38 more
- 2018Sanity Checks for Saliency Mapscited by 1,156
- 2020Beyond Accuracy: Behavioral Testing of NLP models with CheckListcited by 1,006
- 2019Attention is not Explanationcited by 929
- 2019Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailmentcited by 671
- 2019Explaining Machine Learning Classifiers through Diverse Counterfactual Explanationscited by 614
- 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 599
- 2019ERASER: A Benchmark to Evaluate Rationalized NLP Modelscited by 530
- 2020BERT-ATTACK: Adversarial Attack Against BERT Using BERTcited by 491
- 2018Universal Sentence Encodercited by 489
- 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
- 2018A Benchmark for Interpretability Methods in Deep Neural Networkscited by 476
- 2017Interpretation of Neural Networks is Fragilecited by 444
- 2018Generating Natural Language Adversarial Examplescited by 424
- 2015Evaluating the visualization of what a Deep Neural Network has learnedcited by 419
- 2020TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classificationcited by 417
- 2020HateXplain: A Benchmark Dataset for Explainable Hate Speech Detectioncited by 407
- 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 381
- 2018Black-box Generation of Adversarial Text Sequences to Evade Deep Learning Classifierscited by 345
- 2020BAE: BERT-based Adversarial Examples for Text Classificationcited by 316
- 2020Calibration of Pre-trained Transformerscited by 315
- 2018On the Robustness of Interpretability Methodscited by 314
- 2020Pretrained Transformers Improve Out-of-Distribution Robustnesscited by 294
- 2019Feature relevance quantification in explainable AI: A causal problemcited by 167
- 2020Formalizing Trust in Artificial Intelligence: Prerequisites, Causes and Goals of Human Trust in AIcited by 167
- 2019Combating Adversarial Misspellings with Robust Word Recognitioncited by 163
- 2021Post-hoc Interpretability for Neural NLP: A Surveycited by 156
- 2018Explaining Image Classifiers by Counterfactual Generationcited by 144
- 2018Pathologies of Neural Models Make Interpretations Difficultcited by 140
- 2022Towards Faithful Model Explanation in NLP: A Surveycited by 130
- 2023Faithfulness Tests for Natural Language Explanationscited by 107
- 2021The Out-of-Distribution Problem in Explainability and Search Methods for Feature Importance Explanationscited by 87
- 2020Aligning Faithful Interpretations with their Social Attributioncited by 79
- 2022Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanationscited by 77
- 2019Make Up Your Mind! Adversarial Generation of Inconsistent Natural Language Explanationscited by 71
- 2020Evaluations and Methods for Explanation through Robustness Analysiscited by 51
- 2023Adversarial attacks and defenses in explainable artificial intelligence: A surveycited by 48
- 2020The Intriguing Relation Between Counterfactual Explanations and Adversarial Examplescited by 44
- 2022Rethinking Attention-Model Explainability through Faithfulness Violation Testcited by 43
50 most cited of 63 All 63 (JSON)
Built on by
Citing papers per year: 2024: 1; 2025: 1.
2 papers.