mlsanityData to 7 Oct 2026

Learn2Weight: Parameter Adaptation against Similar-domain Adversarial Attacks

arXiv 2205.07315 ↗First posted 15 May 2022cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Recent work in black-box adversarial attacks for NLP systems has attracted much attention. Prior black-box attacks assume that attackers can observe output labels from target models based on selected inputs. In this work, inspired by adversarial transferability, we propose a new type of black-box NLP adversarial attack that an attacker can choose a similar domain and transfer the adversarial examples to the target domain and cause poor performance in target model. Based on domain adaptation theory, we then propose a defensive strategy, called Learn2Weight, which trains to predict the weight adjustments for a target model in order to defend against an attack of similar-domain adversarial examples. Using Amazon multi-domain sentiment classification datasets, we empirically show that Learn2Weight is effective against the attack compared to standard black-box defense methods such as adversarial training and defensive distillation. This work contributes to the growing literature on machine learning safety.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 55 of its 80 references that are in the map. First posted on arXiv 15 May 2022.

Builds on

55 of 80 references.

References · 55

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  3. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  4. 2013Intriguing properties of neural networkscited by 5,114
  5. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  6. 2015Domain-Adversarial Training of Neural Networkscited by 3,023
  7. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  8. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  9. 2016A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networkscited by 2,362
  10. 2016Adversarial examples in the physical worldcited by 1,762
  11. 2019Similarity of Neural Network Representations Revisitedcited by 1,695
  12. 2016Understanding deep learning requires rethinking generalizationcited by 1,657
38 more
  1. 2019Theoretically Principled Trade-off between Robustness and Accuracycited by 1,510
  2. 2018ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustnesscited by 1,430
  3. 2022Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference timecited by 1,366
  4. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  5. 2014How transferable are features in deep neural networks?cited by 1,216
  6. 2014Convolutional Neural Networks for Sentence Classificationcited by 1,124
  7. 2019Learning Robust Global Representations by Penalizing Local Predictive Powercited by 1,010
  8. 2016HyperNetworkscited by 1,003
  9. 2018On First-Order Meta-Learning Algorithmscited by 1,003
  10. 2016Adversarial Machine Learning at Scalecited by 934
  11. 2016Practical Black-Box Attacks against Machine Learningcited by 893
  12. 2017Ensemble Adversarial Training: Attacks and Defensescited by 833
  13. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 709
  14. 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
  15. 2018Improving Transferability of Adversarial Examples with Input Diversitycited by 591
  16. 2018Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interferencecited by 511
  17. 2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 439
  18. 2018Generating Natural Language Adversarial Examplescited by 422
  19. 2020What is being transferred in transfer learning?cited by 365
  20. 2019Bridging Theory and Algorithm for Domain Adaptationcited by 332
  21. 2017Defense against Adversarial Attacks Using High-Level Representation Guided Denoisercited by 329
  22. 2018Adversarially Robust Generalization Requires More Datacited by 318
  23. 2019Continual learning with hypernetworkscited by 267
  24. 2020Training independent subnetworks for robust predictioncited by 177
  25. 2019Certified Robustness to Adversarial Word Substitutionscited by 164
  26. 2020BREEDS: Benchmarks for Subpopulation Shiftcited by 155
  27. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 147
  28. 2018Is Robustness the Cost of Accuracy? -- A Comprehensive Study on the Robustness of 18 Deep Image Classification Modelscited by 135
  29. 2019Adversarial Training Can Hurt Generalizationcited by 131
  30. 2019Improving Black-box Adversarial Attacks with a Transfer-based Priorcited by 121
  31. 2021Toward Understanding the Feature Learning Process of Self-supervised Contrastive Learningcited by 119
  32. 2016Crafting Adversarial Input Sequences for Recurrent Neural Networkscited by 102
  33. 2019Cross-Domain Transferability of Adversarial Perturbationscited by 92
  34. 2019Wasserstein Adversarial Examples via Projected Sinkhorn Iterationscited by 89
  35. 2021Learning Neural Network Subspacescited by 82
  36. 2017Learning to Segment Every Thingcited by 64
  37. 2019The Limitations of Adversarial Training and the Blind-Spot Attackcited by 63
  38. 2019Interpolated Adversarial Training: Achieving Robust Neural Networks without Sacrificing Too Much Accuracycited by 49

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year

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

5 papers.

  1. 2023CT-GAT: Cross-Task Generative Adversarial Attack based on Transferabilitycited by 6
  2. 2025Evaluating and Improving Robustness in Large Language Models: A Survey and Future Directionscited by 5
  3. 2022Interpolating Compressed Parameter Subspacescited by 3
  4. 2022Cross-Reality Re-Rendering: Manipulating between Digital and Physical Realitiescited by 0
  5. 2023Projected Subnetworks Scale Adaptationcited by 0