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 175 篇，第 3 / 4 页

  1. ICLR 2025Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
  2. ICML 2025A Certified Unlearning Approach without Access to Source Data
  3. ICML 2025A Closer Look at Backdoor Attacks on CLIP
  4. ICML 2025Auditing Prompt Caching in Language Model APIs
  5. ICML 2025Black-Box Adversarial Attacks on LLM-Based Code Completion
  6. ICML 2025Connecting Thompson Sampling and UCB: Towards More Efficient Trade-offs Between Privacy and Regret
  7. ICML 2025De-AntiFake: Rethinking the Protective Perturbations Against Voice Cloning Attacks
  8. ICML 2025Differential Privacy Guarantees of Markov Chain Monte Carlo Algorithms
  9. ICML 2025EncryptedLLM: Privacy-Preserving Large Language Model Inference via GPU-Accelerated Fully Homomorphic Encryption
  10. ICML 2025Flexible, Efficient, and Stable Adversarial Attacks on Machine Unlearning
  11. ICML 2025Fragments to Facts: Partial-Information Fragment Inference from LLMs
  12. ICML 2025HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks
  13. ICML 2025ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks
  14. ICML 2025Improving the Variance of Differentially Private Randomized Experiments through Clustering
  15. ICML 2025Interpreting the Repeated Token Phenomenon in Large Language Models
  16. ICML 2025Language Models May Verbatim Complete Text They Were Not Explicitly Trained On
  17. ICML 2025Leveraging Per-Instance Privacy for Machine Unlearning
  18. ICML 2025MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents
  19. ICML 2025On Differential Privacy for Adaptively Solving Search Problems via Sketching
  20. ICML 2025Position: Contextual Integrity is Inadequately Applied to Language Models
  21. ICML 2025Privacy Attacks on Image AutoRegressive Models
  22. ICML 2025Private Lossless Multiple Release
  23. ICML 2025Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks
  24. ICML 2025Textual Unlearning Gives a False Sense of Unlearning
  25. ICML 2025The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
  26. ICML 2025The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
  27. ICML 2025The Role of Randomness in Stability
  28. ICML 2025Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models
  29. ICML 2025Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
  30. ICML 2025Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning
  31. NeurIPS 2025A Reliable Cryptographic Framework for Empirical Machine Unlearning Evaluation
  32. NeurIPS 2025A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers
  33. NeurIPS 2025Attack by Yourself: Effective and Unnoticeable Multi-Category Graph Backdoor Attacks with Subgraph Triggers Pool
  34. NeurIPS 2025Blackbox Model Provenance via Palimpsestic Membership Inference
  35. NeurIPS 2025Blameless Users in a Clean Room: Defining Copyright Protection for Generative Models
  36. NeurIPS 2025Competitive Advantage Attacks to Decentralized Federated Learning
  37. NeurIPS 2025Consensus-Robust Transfer Attacks via Parameter and Representation Perturbations
  38. NeurIPS 2025Convex Approximation of Two-Layer ReLU Networks for Hidden State Differential Privacy
  39. NeurIPS 2025Deep Learning with Plausible Deniability
  40. NeurIPS 2025Distributional Adversarial Attacks and Training in Deep Hedging
  41. NeurIPS 2025Dual-Flow: Transferable Multi-Target, Instance-Agnostic Attacks via $\textit{In-the-wild}$ Cascading Flow Optimization
  42. NeurIPS 2025Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients
  43. NeurIPS 2025Enhancing Diffusion-based Unrestricted Adversarial Attacks via Adversary Preferences Alignment
  44. NeurIPS 2025Exploring the limits of strong membership inference attacks on large language models
  45. NeurIPS 2025FracFace: Breaking The Visual Clues—Fractal-Based Privacy-Preserving Face Recognition
  46. NeurIPS 2025From Counterfactuals to Trees: Competitive Analysis of Model Extraction Attacks
  47. NeurIPS 2025From Dormant to Deleted: Tamper-Resistant Unlearning Through Weight-Space Regularization
  48. NeurIPS 2025Harnessing the Universal Geometry of Embeddings
  49. NeurIPS 2025LOMIA: Label-Only Membership Inference Attacks against Pre-trained Large Vision-Language Models
  50. NeurIPS 2025LoSplit: Loss-Guided Dynamic Split for Training-Time Defense Against Graph Backdoor Attacks

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