隐私攻击论文趋势 2023–2026

主要结论

• 隐私攻击论文从 2023 年的 124 篇增长到 2026 年的 219 篇，三年间扩大到约 1.8 倍。
• ICLR 是发隐私攻击论文最多的会议，2023–2026 年共 262 篇，占四个会议总量的 38%。
• 交叉方向里涨得最快的是对齐与安全：同时属于两个主题的论文从 2023 年的 2 篇增至 2026 年的 80 篇。
• 在「联邦与隐私」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的安全聚合是 1.8 倍，隐私攻击是 1.8 倍。

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

• NeurIPS 2026A Private Empirical Defense Against Privacy Audits
• ICML 2026A Statistical Framework for Analyzing Specification Resistance to Learnware-Inversion Risks
• ICLR 2026(Token-Level) \textbf{InfoRMIA}: Stronger Membership Inference and Privacy Assessment for LLMs
• CVPR 2026Out of Sight Out of Track Adversarial Attacks on Propagation-based
• NeurIPS 2025A Reliable Cryptographic Framework for Empirical Machine Unlearning Evaluation
• ICML 2025A Certified Unlearning Approach without Access to Source Data
• ICLR 2025Adversarial Attacks on Data Attribution
• CVPR 2025Adv-CPG: A Customized Portrait Generation Framework with Facial Adversarial Attacks
• NeurIPS 2024Reproducibility Study on Adversarial Attacks Against Robust Transformer Trackers
• ICML 2024Collective Certified Robustness against Graph Injection Attacks
• ICLR 2024Be Careful What You Smooth For: Label Smoothing Can Be a Privacy Shield but Also a Catalyst for Model Inversion Attacks
• CVPR 2024Efficient Privacy-Preserving Visual Localization Using 3D Ray Clouds
• NeurIPS 2023Bounding the Invertibility of Privacy-preserving Instance Encoding using Fisher Information
• ICML 2023Data Poisoning Attacks Against Multimodal Encoders
• ICLR 2023Adversarial Attacks on Adversarial Bandits
• CVPR 2023Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders
• NeurIPS 2026Adaptive Target-Charging with Privacy Filters and Individual Accounting
• ICML 2026How does Bayesian Sampling help Membership Inference Attacks?
• ICLR 2026Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
• CVPR 2026Revisiting Geometric Obfuscation with Dual Convergent Lines for Privacy-Preserving Image

全部隐私攻击论文

共 138 篇，第 2 / 3 页

51. ICML 2025Improving the Variance of Differentially Private Randomized Experiments through Clustering
52. ICML 2025Interpreting the Repeated Token Phenomenon in Large Language Models
53. ICML 2025Language Models May Verbatim Complete Text They Were Not Explicitly Trained On
54. ICML 2025Leveraging Per-Instance Privacy for Machine Unlearning
55. ICML 2025MELON: Provable Defense Against Indirect Prompt Injection Attacks in AI Agents
56. ICML 2025On Differential Privacy for Adaptively Solving Search Problems via Sketching
57. ICML 2025Position: Contextual Integrity is Inadequately Applied to Language Models
58. ICML 2025Privacy Attacks on Image AutoRegressive Models
59. ICML 2025Private Lossless Multiple Release
60. ICML 2025Revealing Weaknesses in Text Watermarking Through Self-Information Rewrite Attacks
61. ICML 2025Textual Unlearning Gives a False Sense of Unlearning
62. ICML 2025The Canary’s Echo: Auditing Privacy Risks of LLM-Generated Synthetic Text
63. ICML 2025The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
64. ICML 2025The Role of Randomness in Stability
65. ICML 2025Theoretically Unmasking Inference Attacks Against LDP-Protected Clients in Federated Vision Models
66. ICML 2025Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond
67. ICML 2025Underestimated Privacy Risks for Minority Populations in Large Language Model Unlearning
68. ICML 2024A Theoretical Analysis of Backdoor Poisoning Attacks in Convolutional Neural Networks
69. ICML 2024Adversarial Attacks on Combinatorial Multi-Armed Bandits
70. ICML 2024AegisFL: Efficient and Flexible Privacy-Preserving Byzantine-Robust Cross-silo Federated Learning
71. ICML 2024BadPart: Unified Black-box Adversarial Patch Attacks against Pixel-wise Regression Tasks
72. ICML 2024Better Safe than Sorry: Pre-training CLIP against Targeted Data Poisoning and Backdoor Attacks
73. ICML 2024Beyond the Calibration Point: Mechanism Comparison in Differential Privacy
74. ICML 2024Collective Certified Robustness against Graph Injection Attacks
75. ICML 2024Data Poisoning Attacks against Conformal Prediction
76. ICML 2024DataFreeShield: Defending Adversarial Attacks without Training Data
77. ICML 2024Exact Soft Analytical Side-Channel Attacks using Tractable Circuits
78. ICML 2024Fast Adversarial Attacks on Language Models In One GPU Minute
79. ICML 2024FedREDefense: Defending against Model Poisoning Attacks for Federated Learning using Model Update Reconstruction Error
80. ICML 2024Improved Bounds for Pure Private Agnostic Learning: Item-Level and User-Level Privacy
81. ICML 2024Improved Communication-Privacy Trade-offs in $L_2$ Mean Estimation under Streaming Differential Privacy
82. ICML 2024Low-Cost High-Power Membership Inference Attacks
83. ICML 2024Mean Estimation in the Add-Remove Model of Differential Privacy
84. ICML 2024Membership Inference Attacks on Diffusion Models via Quantile Regression
85. ICML 2024Mitigating Privacy Risk in Membership Inference by Convex-Concave Loss
86. ICML 2024Neural Collapse meets Differential Privacy: Curious behaviors of NoisyGD with Near-Perfect Representation Learning
87. ICML 2024Privacy Attacks in Decentralized Learning
88. ICML 2024Privacy Backdoors: Stealing Data with Corrupted Pretrained Models
89. ICML 2024Privacy Profiles for Private Selection
90. ICML 2024Privacy-Preserving Data Release Leveraging Optimal Transport and Particle Gradient Descent
91. ICML 2024Privacy-Preserving Embedding via Look-up Table Evaluation with Fully Homomorphic Encryption
92. ICML 2024Private Truly-Everlasting Robust-Prediction
93. ICML 2024Provable Privacy with Non-Private Pre-Processing
94. ICML 2024Reducing Item Discrepancy via Differentially Private Robust Embedding Alignment for Privacy-Preserving Cross Domain Recommendation
95. ICML 2024Rényi Pufferfish Privacy: General Additive Noise Mechanisms and Privacy Amplification by Iteration via Shift Reduction Lemmas
96. ICML 2024Rethinking DP-SGD in Discrete Domain: Exploring Logistic Distribution in the Realm of signSGD
97. ICML 2024Revisiting Character-level Adversarial Attacks for Language Models
98. ICML 2024Shifted Interpolation for Differential Privacy
99. ICML 2024SignSGD with Federated Defense: Harnessing Adversarial Attacks through Gradient Sign Decoding
100. ICML 2024Split-and-Denoise: Protect large language model inference with local differential privacy

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。