隐私攻击论文趋势 2023–2026

成员推断等

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 698 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

隐私攻击论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202346242925124
202463534222180
202545832918175
2026551023824219

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

全部隐私攻击论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 698 篇，第 7 / 14 页

  1. ICLR 2025Scalable Extraction of Training Data from Aligned, Production Language Models
  2. ICLR 2025SelectFormer in Data Markets: Privacy-Preserving and Efficient Data Selection for Transformers with Multi-Party Computation
  3. ICLR 2025SimPER: A Minimalist Approach to Preference Alignment without Hyperparameters
  4. ICLR 2025Singular Subspace Perturbation Bounds via Rectangular Random Matrix Diffusions
  5. ICLR 2025Stealthy Shield Defense: A Conditional Mutual Information-Based Approach against Black-Box Model Inversion Attacks
  6. ICLR 2025Stochastic Bandits Robust to Adversarial Attacks
  7. ICLR 2025Support is All You Need for Certified VAE Training
  8. ICLR 2025Tamper-Resistant Safeguards for Open-Weight LLMs
  9. ICLR 2025Targeted Attack Improves Protection against Unauthorized Diffusion Customization
  10. ICLR 2025The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD
  11. ICLR 2025Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model
  12. ICLR 2025To Tackle Adversarial Transferability: A Novel Ensemble Training Method with Fourier Transformation
  13. ICLR 2025Towards Certification of Uncertainty Calibration under Adversarial Attacks
  14. ICLR 2025Tracking the Copyright of Large Vision-Language Models through Parameter Learning Adversarial Images
  15. ICLR 2025Training Robust Ensembles Requires Rethinking Lipschitz Continuity
  16. ICLR 2025Understanding and Enhancing the Transferability of Jailbreaking Attacks
  17. ICLR 2025Unlearn and Burn: Adversarial Machine Unlearning Requests Destroy Model Accuracy
  18. ICLR 2025Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning
  19. ICLR 2025Variational Best-of-N Alignment
  20. ICLR 2025Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
  21. ICML 2025A Certified Unlearning Approach without Access to Source Data
  22. ICML 2025A Closer Look at Backdoor Attacks on CLIP
  23. ICML 2025Auditing Prompt Caching in Language Model APIs
  24. ICML 2025Black-Box Adversarial Attacks on LLM-Based Code Completion
  25. ICML 2025Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret
  26. ICML 2025De-AntiFake: Rethinking the Protective Perturbations Against Voice Cloning Attacks
  27. ICML 2025Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms
  28. ICML 2025EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption
  29. ICML 2025Flexible, Efficient, and Stable Adversarial Attacks on Machine Unlearning
  30. ICML 2025Fragments to Facts: Partial-Information Fragment Inference from LLMs
  31. ICML 2025HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks
  32. ICML 2025ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks
  33. ICML 2025Improving the Variance of Differentially Private Randomized Experiments through Clustering
  34. ICML 2025Interpreting the Repeated Token Phenomenon in Large Language Models
  35. ICML 2025Language Models May Verbatim Complete Text They Were Not Explicitly Trained On
  36. ICML 2025Leveraging Per-Instance Privacy for Machine Unlearning
  37. ICML 2025MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents
  38. ICML 2025On Differential Privacy for Adaptively Solving Search Problems via Sketching
  39. ICML 2025Position: Contextual Integrity is Inadequately Applied to Language Models
  40. ICML 2025Privacy Attacks on Image AutoRegressive Models
  41. ICML 2025Private Lossless Multiple Release
  42. ICML 2025Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks
  43. ICML 2025Textual Unlearning Gives a False Sense of Unlearning
  44. ICML 2025The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
  45. ICML 2025The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
  46. ICML 2025The Role of Randomness in Stability
  47. ICML 2025Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models
  48. ICML 2025Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
  49. ICML 2025Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning
  50. NeurIPS 2025A Reliable Cryptographic Framework for Empirical Machine Unlearning Evaluation

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 查隐私攻击在 ICLR 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": "privacy-attacks"}

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隐私攻击论文？
共 698 篇，按年份分别是 2023 年 124 篇、2024 年 180 篇、2025 年 175 篇、2026 年 219 篇。
NeurIPS 2026 有多少篇隐私攻击论文？
NeurIPS 2026 有 55 篇隐私攻击论文，上一届是 45 篇。
ICLR 2026 有多少篇隐私攻击论文？
ICLR 2026 有 102 篇隐私攻击论文，上一届是 83 篇。
ICML 2026 有多少篇隐私攻击论文？
ICML 2026 有 38 篇隐私攻击论文，上一届是 29 篇。
CVPR 2026 有多少篇隐私攻击论文？
CVPR 2026 有 24 篇隐私攻击论文，上一届是 18 篇。
哪个会议的隐私攻击论文最多？
ICLR，2023–2026 年共 262 篇。

「联邦与隐私」下的其他主题