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 698 篇，第 13 / 14 页

601. ICLR 2023Adversarial Attacks on Adversarial Bandits
602. ICLR 2023Canary in a Coalmine: Better Membership Inference with Ensembled Adversarial Queries
603. ICLR 2023CANIFE: Crafting Canaries for Empirical Privacy Measurement in Federated Learning
604. ICLR 2023Decepticons: Corrupted Transformers Breach Privacy in Federated Learning for Language Models
605. ICLR 2023Differentially Private $L_2$-Heavy Hitters in the Sliding Window Model
606. ICLR 2023Disparate Impact in Differential Privacy from Gradient Misalignment
607. ICLR 2023Distributed Differential Privacy in Multi-Armed Bandits
608. ICLR 2023Easy Differentially Private Linear Regression
609. ICLR 2023Effective passive membership inference attacks in federated learning against overparameterized models
610. ICLR 2023Exploring the Limits of Differentially Private Deep Learning with Group-wise Clipping
611. ICLR 2023Extracting Robust Models with Uncertain Examples
612. ICLR 2023Few-shot Backdoor Attacks via Neural Tangent Kernels
613. ICLR 2023Incompatibility Clustering as a Defense Against Backdoor Poisoning Attacks
614. ICLR 2023Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning
615. ICLR 2023Individual Privacy Accounting with Gaussian Differential Privacy
616. ICLR 2023Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial Examples
617. ICLR 2023Measuring Forgetting of Memorized Training Examples
618. ICLR 2023MPCFORMER: FAST, PERFORMANT AND PRIVATE TRANSFORMER INFERENCE WITH MPC
619. ICLR 2023Panning for Gold in Federated Learning: Targeted Text Extraction under Arbitrarily Large-Scale Aggregation
620. ICLR 2023Quantifying Memorization Across Neural Language Models
621. ICLR 2023Regression with Label Differential Privacy
622. ICLR 2023Robust Multivariate Time-Series Forecasting: Adversarial Attacks and Defense Mechanisms
623. ICLR 2023Share Your Representation Only: Guaranteed Improvement of the Privacy-Utility Tradeoff in Federated Learning
624. ICML 2023Adversarial Parameter Attack on Deep Neural Networks
625. ICML 2023Algorithms for bounding contribution for histogram estimation under user-level privacy
626. ICML 2023Are Diffusion Models Vulnerable to Membership Inference Attacks?
627. ICML 2023Bayesian Estimation of Differential Privacy
628. ICML 2023Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning Using Independent Component Analysis
629. ICML 2023Concurrent Shuffle Differential Privacy Under Continual Observation
630. ICML 2023Data Poisoning Attacks Against Multimodal Encoders
631. ICML 2023Demystifying Uneven Vulnerability of Link Stealing Attacks against Graph Neural Networks
632. ICML 2023Differential Privacy has Bounded Impact on Fairness in Classification
633. ICML 2023Differential Privacy, Linguistic Fairness, and Training Data Influence: Impossibility and Possibility Theorems for Multilingual Language Models
634. ICML 2023Effectively Using Public Data in Privacy Preserving Machine Learning
635. ICML 2023Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning Attacks
636. ICML 2023Federated Linear Contextual Bandits with User-level Differential Privacy
637. ICML 2023From Robustness to Privacy and Back
638. ICML 2023Fully-Adaptive Composition in Differential Privacy
639. ICML 2023Graph Contrastive Backdoor Attacks
640. ICML 2023GuardHFL: Privacy Guardian for Heterogeneous Federated Learning
641. ICML 2023HETAL: Efficient Privacy-preserving Transfer Learning with Homomorphic Encryption
642. ICML 2023MultiRobustBench: Benchmarking Robustness Against Multiple Attacks
643. ICML 2023Preprocessors Matter! Realistic Decision-Based Attacks on Machine Learning Systems
644. ICML 2023Privacy-Aware Compression for Federated Learning Through Numerical Mechanism Design
645. ICML 2023Rethinking Backdoor Attacks
646. ICML 2023Run-off Election: Improved Provable Defense against Data Poisoning Attacks
647. ICML 2023Sketching Meets Differential Privacy: Fast Algorithm for Dynamic Kronecker Projection Maintenance
648. ICML 2023Streaming Submodular Maximization with Differential Privacy
649. ICML 2023The Price of Differential Privacy under Continual Observation
650. ICML 2023The Saddle-Point Method in 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。