mlsanityData to 14 Oct 2026

Efficient Explanations from Empirical Explainers

arXiv 2103.15429 ↗First posted 29 Mar 2021cs.LG cs.CLcited by 9

Not on the map. Why?

Abstract (arXiv)

Amid a discussion about Green AI in which we see explainability neglected, we explore the possibility to efficiently approximate computationally expensive explainers. To this end, we propose feature attribution modelling with Empirical Explainers. Empirical Explainers learn from data to predict the attribution maps of expensive explainers. We train and test Empirical Explainers in the language domain and find that they model their expensive counterparts surprisingly well, at a fraction of the cost. They could thus mitigate the computational burden of neural explanations significantly, in applications that tolerate an approximation error.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 33 of its 41 references that are in the map. First posted on arXiv 29 Mar 2021.

Builds on

33 of 41 references.

References · 33

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  5. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  7. 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,811
  8. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  9. 2013Visualizing and Understanding Convolutional Networkscited by 2,290
  10. 2015A large annotated corpus for learning natural language inferencecited by 1,847
  11. 2019Energy and Policy Considerations for Deep Learning in NLPcited by 1,254
  12. 2017SmoothGrad: removing noise by adding noisecited by 1,109
21 more
  1. 2019Green AIcited by 627
  2. 2020Towards Faithfully Interpretable NLP Systems: How should we define and evaluate faithfulness?cited by 599
  3. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 547
  4. 2018Towards Robust Interpretability with Self-Explaining Neural Networkscited by 505
  5. 2019PAWS: Paraphrase Adversaries from Word Scramblingcited by 402
  6. 2016European Union regulations on algorithmic decision-making and a "right to explanation"cited by 344
  7. 2017Right for the Right Reasons: Training Differentiable Models by Constraining their Explanationscited by 294
  8. 2020Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learningcited by 249
  9. 2020Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regressioncited by 154
  10. 2019The Shapley Taylor Interaction Indexcited by 143
  11. 2020Making deep neural networks right for the right scientific reasons by interacting with their explanationscited by 142
  12. 2020Restricting the Flow: Information Bottlenecks for Attributioncited by 110
  13. 2018L-Shapley and C-Shapley: Efficient Model Interpretation for Structured Datacited by 100
  14. 2020Evaluating Explanations: How much do explanations from the teacher aid students?cited by 94
  15. 2019Incorporating Priors with Feature Attribution on Text Classificationcited by 68
  16. 2021SelfExplain: A Self-Explaining Architecture for Neural Text Classifierscited by 44
  17. 2019Improving performance of deep learning models with axiomatic attribution priors and expected gradientscited by 32
  18. 2020Investigating Saturation Effects in Integrated Gradientscited by 22
  19. 2020Considering Likelihood in NLP Classification Explanations with Occlusion and Language Modelingcited by 11
  20. 2020Pattern-Guided Integrated Gradientscited by 1
  21. 2021Bandits for Learning to Explain from Explanationscited by 1

Built on by

Cited by year

Citing papers per year: 2021: 2; 2022: 0; 2023: 3; 2024: 3; 2025: 1.

9 papers.

  1. 2021UNIREX: A Unified Learning Framework for Language Model Rationale Extractioncited by 32
  2. 2024Stochastic Amortization: A Unified Approach to Accelerate Feature and Data Attributioncited by 21
  3. 2023Efficient XAI Techniques: A Taxonomic Surveycited by 14
  4. 2025Learning to Attribute with Attentioncited by 13
  5. 2021Thermostat: A Large Collection of NLP Model Explanations and Analysis Toolscited by 11
  6. 2024SPES: Spectrogram Perturbation for Explainable Speech-to-Text Generationcited by 6
  7. 2023REFER: An End-to-end Rationale Extraction Framework for Explanation Regularizationcited by 4
  8. 2023Fast Shapley Value Estimation: A Unified Approachcited by 2
  9. 2024Selective Explanationscited by 2