mlsanityData to 14 Oct 2026

Smaller is Better: Enhancing Transparency in Vehicle AI Systems via Pruning

arXiv 2509.20148 ↗First posted 24 Sep 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

Connected and autonomous vehicles continue to heavily rely on AI systems, where transparency and security are critical for trust and operational safety. Post-hoc explanations provide transparency to these black-box like AI models but the quality and reliability of these explanations is often questioned due to inconsistencies and lack of faithfulness in representing model decisions. This paper systematically examines the impact of three widely used training approaches, namely natural training, adversarial training, and pruning, affect the quality of post-hoc explanations for traffic sign classifiers. Through extensive empirical evaluation, we demonstrate that pruning significantly enhances the comprehensibility and faithfulness of explanations (using saliency maps). Our findings reveal that pruning not only improves model efficiency but also enforces sparsity in learned representation, leading to more interpretable and reliable decisions. Additionally, these insights suggest that pruning is a promising strategy for developing transparent deep learning models, especially in resource-constrained vehicular AI systems.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 37 of its 47 references that are in the map. First posted on arXiv 24 Sep 2025.

Builds on

37 of 47 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2014Explaining and Harnessing Adversarial Examplescited by 7,364
  4. 2016Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localizationcited by 6,733
  5. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  6. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,415
  7. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  8. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  9. 2013Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Mapscited by 2,371
  10. 2015Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Codingcited by 2,329
  11. 2018The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networkscited by 2,262
  12. 2015Learning both Weights and Connections for Efficient Neural Networkscited by 2,031
25 more
  1. 2017Learning Important Features Through Propagating Activation Differencescited by 1,458
  2. 2018Sanity Checks for Saliency Mapscited by 1,156
  3. 2017SmoothGrad: removing noise by adding noisecited by 1,109
  4. 2014Striving for Simplicity: The All Convolutional Netcited by 990
  5. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 838
  6. 2020What is the State of Neural Network Pruning?cited by 583
  7. 2020Picking Winning Tickets Before Training by Preserving Gradient Flowcited by 471
  8. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  9. 2017The (Un)reliability of saliency methodscited by 333
  10. 2018Soft Filter Pruning for Accelerating Deep Convolutional Neural Networkscited by 315
  11. 2023A Survey on Deep Neural Network Pruning-Taxonomy, Comparison, Analysis, and Recommendationscited by 202
  12. 2019A Survey of Deep Learning Applications to Autonomous Vehicle Controlcited by 106
  13. 2022A Consistent and Efficient Evaluation Strategy for Attribution Methodscited by 102
  14. 2019On the Connection Between Adversarial Robustness and Saliency Map Interpretabilitycited by 80
  15. 2017More is Less: A More Complicated Network with Less Inference Complexitycited by 66
  16. 2019ML-LOO: Detecting Adversarial Examples with Feature Attributioncited by 48
  17. 2021NoiseGrad: Enhancing Explanations by Introducing Stochasticity to Model Weightscited by 28
  18. 2019Improving Feature Attribution through Input-specific Network Pruningcited by 7
  19. 2022SoK: Modeling Explainability in Security Analytics for Interpretability, Trustworthiness, and Usabilitycited by 7
  20. 2024Structured Gradient-based Interpretations via Norm-Regularized Adversarial Trainingcited by 7
  21. 2023Interpretability is a Kind of Safety: An Interpreter-based Ensemble for Adversary Defensecited by 6
  22. 2017Interpreting Convolutional Neural Networks Through Compressioncited by 5
  23. 2023Less is More: The Influence of Pruning on the Explainability of CNNscited by 5
  24. 2024PASA: Attack Agnostic Unsupervised Adversarial Detection using Prediction & Attribution Sensitivity Analysiscited by 3
  25. 2025Do Sparse Subnetworks Exhibit Cognitively Aligned Attention? Effects of Pruning on Saliency Map Fidelity, Sparsity, and Concept Coherencecited by 1

Built on by

Not cited yet.