对抗鲁棒论文趋势 2023–2026

对抗样本与防御

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 1,291 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

对抗鲁棒论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
2023100796074313
2024131707360334
2025110888544327
202681949349317

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

全部对抗鲁棒论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 311 篇，第 2 / 7 页

  1. ICML 2026On Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMs
  2. ICML 2026On the Adversarial Robustness of Large Vision-Language Models under Visual Token Compression
  3. ICML 2026On the Existence of Consistent Adversarial Attacks in High-Dimensional Linear Classification
  4. ICML 2026On the Robustness of Langevin Dynamics to Score Function Error
  5. ICML 2026Optimal Attention Temperature Improves the Robustness of In-Context Learning under Distribution Shift in High Dimensions
  6. ICML 2026Posterior Mismatch Matters: Adversarial Training for Long-Tailed Robustness
  7. ICML 2026Proactive Defense Benchmark against Deepfake Generation
  8. ICML 2026Probabilistic Robustness Certificates against Adversarial Attacks
  9. ICML 2026Provably Efficient Policy-Reward Co-Pretraining for Adversarial Imitation Learning
  10. ICML 2026PSMix: Robust Point Cloud Recognition through Spectral Domain Mixing
  11. ICML 2026RADAR: Defending RAG Dynamically against Retrieval Corruption
  12. ICML 2026Regret Minimization With a Crowd of Awakening Experts
  13. ICML 2026Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
  14. ICML 2026Ripple Perturbations Through Structure: Likelihood-Constrained Adversarial Attacks on Heterogeneous Tabular Data
  15. ICML 2026Robust Bayesian Optimisation with Unbounded Corruptions
  16. ICML 2026Robust Linear Dueling Bandits with Post-serving Context under Unknown Delays and Adversarial Corruptions
  17. ICML 2026Robust Vision-Language Models via Manifold-Adversarial Adapters
  18. ICML 2026Robustness of Mixtures of Experts to Feature Noise
  19. ICML 2026Safety-Efficacy Trade Off: Robustness against Data-Poisoning
  20. ICML 2026Scalable GANs with Transformers
  21. ICML 2026Security--Fidelity Tradeoffs: The Hidden Cost of Prompt Injection Defense
  22. ICML 2026Segment Anything with Robust Uncertainty-Accuracy Correlation
  23. ICML 2026Self-Calibrated Consistency can Fight Back for Adversarial Robustness in Vision-Language Models
  24. ICML 2026Semantic Router: On the Feasibility of Hijacking MLLMs via a Single Adversarial Perturbation
  25. ICML 2026Ski Rental with Distributional Predictions of Unknown Quality
  26. ICML 2026SORA: Free Second-Order Attacks in Fast Adversarial Training
  27. ICML 2026SS-TPT: Stability and Suitability-Guided Test-Time Prompt Tuning for Adversarially Robust Vision-Language Models
  28. ICML 2026Statistically Undetectable Backdoors in Deep Neural Networks
  29. ICML 2026Steering Beyond the Support: Adversarial Training on Unsupervised Jailbroken Activation Simulation
  30. ICML 2026Stochastic Order Learning: An Approach to Rank Estimation Using Noisy Data
  31. ICML 2026Suppress and Diversify: Refining Robust Pathways for Corruption Robustness
  32. ICML 2026SWE-ABS: Adversarial Benchmark Strengthening Exposes Inflated Success Rates on Test-based Benchmark
  33. ICML 2026The Latent Guardian: Defending Collaborative Perception via Feature-Level Consistency Verification
  34. ICML 2026Toward Understanding Adversarial Distillation: Why Robust Teachers Fail
  35. ICML 2026Towards Fine-Grained Robustness: Attention-Guided Test-Time Prompt Tuning for Vision-Language Models
  36. ICML 2026Towards Optimal Robustness in Learning-Augmented Paging
  37. ICML 2026Towards Understanding Generalization of Federated Adversarial Learning: Perspective of Algorithmic Stability
  38. ICML 2026Training with Honeypots: Reshaping How LLMs Fail Under Adversarial Attacks
  39. ICML 2026Training-Free Adversarial Robustness in Computational MRI
  40. ICML 2026Two Modalities Are Better Than One: Efficient Adversarial Purification via Multimodal Diffusion Models
  41. ICML 2026Unifying Adversarial Robustness and Training Across Text Scoring Models
  42. ICML 2026When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPs
  43. ICML 2026You Don't Protect if You Don't Expect: Breaking the Key Assumption behind CLIP's Test-Time Defenses
  44. ICML 2025Adapting to Evolving Adversaries with Regularized Continual Robust Training
  45. ICML 2025Adaptive Median Smoothing: Adversarial Defense for Unlearned Text-to-Image Diffusion Models at Inference Time
  46. ICML 2025Adversarial Combinatorial Semi-bandits with Graph Feedback
  47. ICML 2025Adversarial Inputs for Linear Algebra Backends
  48. ICML 2025Adversarial Perturbations Are Formed by Iteratively Learning Linear Combinations of the Right Singular Vectors of the Adversarial Jacobian
  49. ICML 2025Adversarial Reasoning at Jailbreaking Time
  50. ICML 2025Adversarial Robust Generalization of Graph Neural Networks

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查对抗鲁棒在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "adversarial-robustness"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇对抗鲁棒论文？
共 1,291 篇，按年份分别是 2023 年 313 篇、2024 年 334 篇、2025 年 327 篇、2026 年 317 篇。
NeurIPS 2026 有多少篇对抗鲁棒论文？
NeurIPS 2026 有 81 篇对抗鲁棒论文，上一届是 110 篇。
ICLR 2026 有多少篇对抗鲁棒论文？
ICLR 2026 有 94 篇对抗鲁棒论文，上一届是 88 篇。
ICML 2026 有多少篇对抗鲁棒论文？
ICML 2026 有 93 篇对抗鲁棒论文，上一届是 85 篇。
CVPR 2026 有多少篇对抗鲁棒论文？
CVPR 2026 有 49 篇对抗鲁棒论文，上一届是 44 篇。
哪个会议的对抗鲁棒论文最多？
NeurIPS，2023–2026 年共 422 篇。

「安全与可信」下的其他主题