Research map

Adversarial robustness and security: research map

Adversarial attacks and defenses, backdoors and poisoning.

2,947 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026), in 4 topics. Within all five venues, its share held steady from 2.9% in 2023–24 to 2.2% in 2025–26 (778 → 982 papers at ICML, NeurIPS, CVPR and AAAI, the venues with data for all four years).

Explore Adversarial robustness and security in the interactive map

Topics

• Certified and empirical robustness · 1,156 papers
  Robustness-accuracy trade-off analysis, Attack evaluation and defense critique, Adversarial training refinements
• Attacks on text and graphs · 1,088 papers
  Defenses against universal perturbations, Attacks and defenses in sequential settings, Diffusion-based adversarial purification
• Black-box and transfer attacks · 383 papers
  Robustness benchmarks and defenses, Query-efficient black-box attacks, Data-free substitute training
• Backdoor and poisoning · 320 papers
  Trigger-based backdoor attacks, Backdoors in large multimodal models, Backdoors in federated learning

Most cited papers in Adversarial robustness and security

• DeepFool: A Simple and Accurate Method to Fool Deep Neural Networks (CVPR 2016 · 5,359 citations)
• mixup: Beyond Empirical Risk Minimization (ICLR 2018 · 4,723 citations)
• Boosting Adversarial Attacks With Momentum (CVPR 2018 · 3,047 citations)
• Universal Adversarial Perturbations (CVPR 2017 · 2,731 citations)
• Ensemble Adversarial Training: Attacks and Defenses (ICLR 2018 · 2,314 citations)
• Robust Physical-World Attacks on Deep Learning Visual Classification (CVPR 2018 · 2,168 citations)
• Towards Deep Learning Models Resistant to Adversarial Attacks (ICLR 2018 · 1,508 citations)
• Improving Transferability of Adversarial Examples With Input Diversity (CVPR 2019 · 1,273 citations)
• Theoretically Principled Trade-off between Robustness and Accuracy (ICML 2019 · 1,119 citations)
• Defense Against Adversarial Attacks Using High-Level Representation Guided Denoiser (CVPR 2018 · 945 citations)
• Provable Defenses against Adversarial Examples via the Convex Outer Adversarial Polytope (ICML 2018 · 674 citations)
• Synthesizing Robust Adversarial Examples (ICML 2018 · 667 citations)