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 262 篇，第 1 / 6 页

1. ICLR 2026(Token-Level) \textbf{InfoRMIA}: Stronger Membership Inference and Privacy Assessment for LLMs
2. ICLR 2026A General Framework for Black-Box Attacks Under Cost Asymmetry
3. ICLR 2026A Law of Data Reconstruction for Random Features (And Beyond)
4. ICLR 2026Adaptive Attacks on Trusted Monitors Subvert AI Control Protocols
5. ICLR 2026Adversarial Attacks Already Tell the Answer: Directional Bias-Guided Test-time Defense for Vision-Language Models
6. ICLR 2026Adversarial Déjà Vu: Jailbreak Dictionary Learning for Stronger Generalization to Unseen Attacks
7. ICLR 2026AEGIS: Adversarial Target–Guided Retention-Data-Free Robust Concept Erasure from Diffusion Models
8. ICLR 2026Align to Misalign: Automatic LLM Jailbreak with Meta-Optimized LLM Judges
9. ICLR 2026AlphaSteer: Learning Refusal Steering with Principled Null-Space Constraint
10. ICLR 2026Antibody: Strengthening Defense Against Harmful Fine-Tuning for Large Language Models via Attenuating Harmful Gradient Influence
11. ICLR 2026Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth
12. ICLR 2026Automatic Dialectic Jailbreak: A Framework for Generating Effective Jailbreak Strategies
13. ICLR 2026Benchmarking Empirical Privacy Protection for Adaptations of Large Language Models
14. ICLR 2026Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning
15. ICLR 2026Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
16. ICLR 2026Black-Box Privacy Attacks on Shared Representations in Multitask Learning
17. ICLR 2026Concept-Aware Privacy Mechanisms for Defending Embedding Inversion Attacks
18. ICLR 2026Curation Leaks: Membership Inference Attacks against Data Curation for Machine Learning
19. ICLR 2026Defending against Backdoor Attacks via Module Switching
20. ICLR 2026DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
21. ICLR 2026Disrupting Hierarchical Reasoning: Adversarial Protection for Geographic Privacy in Multimodal Reasoning Models
22. ICLR 2026Don't Shift the Trigger: Robust Gradient Ascent for Backdoor Unlearning
23. ICLR 2026Doxing via the Lens: Revealing Location-related Privacy Leakage on Multi-modal Large Reasoning Models
24. ICLR 2026DRIFT: Divergent Response in Filtered Transformations for Robust Adversarial Defense
25. ICLR 2026Dropping Just a Handful of Preferences Can Change Top Large Language Model Rankings
26. ICLR 2026Dual Randomized Smoothing: Beyond Global Noise Variance
27. ICLR 2026Dual-Space Smoothness for Robust and Balanced LLM Unlearning
28. ICLR 2026Efficient Adversarial Attacks on High-dimensional Offline Bandits
29. ICLR 2026EigenBench: A Comparative Behavioral Measure of Value Alignment
30. ICLR 2026Eliciting Harmful Capabilities by Fine-Tuning on Safeguarded Outputs
31. ICLR 2026EvA: Evolutionary Attacks on Graphs
32. ICLR 2026Exponential-Wrapped Mechanisms: Differential Privacy on Hadamard Manifolds Made Practical
33. ICLR 2026Federated Learning of Quantile Inference under Local Differential Privacy
34. ICLR 2026Federated Learning with Profile Mapping under Distribution Shifts and Drifts
35. ICLR 2026Fine-Grained Iterative Adversarial Attacks with Limited Computation Budget
36. ICLR 2026Fine-Grained Privacy Extraction from Retrieval-Augmented Generation Systems by Exploiting Knowledge Asymmetry
37. ICLR 2026Fingerprinting Deep Neural Networks for Ownership Protection: An Analytical Approach
38. ICLR 2026Ghost in the Cloud: Your Geo-Distributed Large Language Models Training is Easily Manipulated
39. ICLR 2026GuidedBench: Measuring and Mitigating the Evaluation Discrepancies of In-the-wild LLM Jailbreak Methods
40. ICLR 2026Hubble: a Model Suite to Advance the Study of LLM Memorization
41. ICLR 2026INO-SGD: Addressing Utility Imbalance under Individualized Differential Privacy
42. ICLR 2026Jailbreak Transferability Emerges from Shared Representations
43. ICLR 2026JailbreakLoRA: Your Downloaded LoRA from Sharing Platforms might be Unsafe
44. ICLR 2026JULI: Jailbreak Large Language Models by Self-Introspection
45. ICLR 2026Label Smoothing Improves Machine Unlearning
46. ICLR 2026Language Models are Injective and Hence Invertible
47. ICLR 2026Learnability and Privacy Vulnerability are Entangled in a Few Critical Weights
48. ICLR 2026Learning to Lie: Adversarial Attacks on Human-AI Teams and LLMs
49. ICLR 2026Membership Inference Attacks Against Fine-tuned Diffusion Language Models
50. ICLR 2026Membership Privacy Risks of Sharpness Aware Minimization

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。