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 316 篇，第 5 / 7 页

  1. ICML 2026Provably Efficient Policy-Reward Co-Pretraining for Adversarial Imitation Learning
  2. ICML 2026PSMix: Robust Point Cloud Recognition through Spectral Domain Mixing
  3. ICML 2026RADAR: Defending RAG Dynamically against Retrieval Corruption
  4. ICML 2026Regret Minimization With a Crowd of Awakening Experts
  5. ICML 2026Reward Under Attack: Analyzing the Robustness and Hackability of Process Reward Models
  6. ICML 2026Ripple Perturbations Through Structure: Likelihood-Constrained Adversarial Attacks on Heterogeneous Tabular Data
  7. ICML 2026Robust Bayesian Optimisation with Unbounded Corruptions
  8. ICML 2026Robust Linear Dueling Bandits with Post-serving Context under Unknown Delays and Adversarial Corruptions
  9. ICML 2026Robust Vision-Language Models via Manifold-Adversarial Adapters
  10. ICML 2026Robustness of Mixtures of Experts to Feature Noise
  11. ICML 2026Safety-Efficacy Trade Off: Robustness against Data-Poisoning
  12. ICML 2026Scalable GANs with Transformers
  13. ICML 2026Security--Fidelity Tradeoffs: The Hidden Cost of Prompt Injection Defense
  14. ICML 2026Segment Anything with Robust Uncertainty-Accuracy Correlation
  15. ICML 2026Self-Calibrated Consistency can Fight Back for Adversarial Robustness in Vision-Language Models
  16. ICML 2026Semantic Router: On the Feasibility of Hijacking MLLMs via a Single Adversarial Perturbation
  17. ICML 2026Ski Rental with Distributional Predictions of Unknown Quality
  18. ICML 2026SORA: Free Second-Order Attacks in Fast Adversarial Training
  19. ICML 2026SS-TPT: Stability and Suitability-Guided Test-Time Prompt Tuning for Adversarially Robust Vision-Language Models
  20. ICML 2026Statistically Undetectable Backdoors in Deep Neural Networks
  21. ICML 2026Steering Beyond the Support: Adversarial Training on Unsupervised Jailbroken Activation Simulation
  22. ICML 2026Stochastic Order Learning: An Approach to Rank Estimation Using Noisy Data
  23. ICML 2026Suppress and Diversify: Refining Robust Pathways for Corruption Robustness
  24. ICML 2026SWE-ABS: Adversarial Benchmark Strengthening Exposes Inflated Success Rates on Test-based Benchmark
  25. ICML 2026The Latent Guardian: Defending Collaborative Perception via Feature-Level Consistency Verification
  26. ICML 2026Toward Understanding Adversarial Distillation: Why Robust Teachers Fail
  27. ICML 2026Towards Fine-Grained Robustness: Attention-Guided Test-Time Prompt Tuning for Vision-Language Models
  28. ICML 2026Towards Optimal Robustness in Learning-Augmented Paging
  29. ICML 2026Towards Understanding Generalization of Federated Adversarial Learning: Perspective of Algorithmic Stability
  30. ICML 2026Training with Honeypots: Reshaping How LLMs Fail Under Adversarial Attacks
  31. ICML 2026Training-Free Adversarial Robustness in Computational MRI
  32. ICML 2026Two Modalities Are Better Than One: Efficient Adversarial Purification via Multimodal Diffusion Models
  33. ICML 2026Unifying Adversarial Robustness and Training Across Text Scoring Models
  34. ICML 2026When Can You Poison Rewards? A Tight Characterization of Reward Poisoning in Linear MDPs
  35. ICML 2026You Don't Protect if You Don't Expect: Breaking the Key Assumption behind CLIP's Test-Time Defenses
  36. NeurIPS 2026A Little Robustness Is All You Need: Leveraging Predictions for Contextual Optimization
  37. NeurIPS 2026A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA)
  38. NeurIPS 2026A Regularization-Based Approach to Public Belief State Search for Adversarial Games
  39. NeurIPS 2026A Retained-Signal Interface for LLM Watermark Robustness under Paraphrase
  40. NeurIPS 2026A Simple Class-Agnostic Approach to Enhance Fair Adversarial Training
  41. NeurIPS 2026A Solver-Efficient Neural Adversarial Attack on Subgraph Matching Models
  42. NeurIPS 2026Adversarial Corpus Selection to Attack Subgraph Matching based Graph Retrieval
  43. NeurIPS 2026Adversarial Group Fairness in Contextual Bandits: When Robust is Not Fair
  44. NeurIPS 2026AI Models Can Provably Hide Arbitrary Capabilities
  45. NeurIPS 2026AnchorRep: Defending LLMs Against Cross-Model Adversarial Transfer via Representation Repulsion
  46. NeurIPS 2026Assessing the robustness of heterogeneous treatment effects in survival analysis under informative censoring
  47. NeurIPS 2026BayesAT: Bayes-Guided Progressive Distillation for Semi-Supervised Adversarial Training
  48. NeurIPS 2026Benchmarking Sensor-Fault Robustness in Forecasting
  49. NeurIPS 2026Beyond Correctness: Robustness-Driven Evolutionary Self-Training for Large Language Models
  50. NeurIPS 2026Beyond Ground Truth: Evaluating Non-Verifiable Reasoning in LLMs through Moral Robustness

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