Social aspects · privacy papers, 2023–2026

Key findings

• Social aspects · privacy went from 21 papers in 2023 to 138 in 2026, roughly 6.6x in three years.
• NeurIPS publishes the most Social aspects · privacy work: 215 papers from 2023 to 2026, 50% of the four-venue total.
• The fastest-growing overlap is with Social aspects · accountability transparency and interpretability: papers tagged with both rose from 0 in 2023 to 93 in 2026.
• Among the 5 topics under Alignment and Responsible AI, it ranks #4 by 2026 paper count; the fastest grower in the group, Social aspects · safety, is up 93x versus 6.6x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026Assessing Per-Sample Membership Inference Vulnerability without Retraining
• ICML 2026Accuracy-First Rényi Differential Privacy and Post-Processing Immunity
• ICLR 2026Mitigating Privacy Risk via Forget Set-Free Unlearning
• CVPR 2026Protego User-Centric Pose-Invariant Privacy Protection Against Face Recognition-Induced Digital Footprint
• NeurIPS 2025A Generalized Binary Tree Mechanism for Private Approximation of All-Pair Shortest Distances
• ICML 2025Breaking the $n^{1.5}$ Additive Error Barrier for Private and Efficient Graph Sparsification via Private Expander Decomposition
• ICLR 2025Machine Unlearning via Simulated Oracle Matching
• CVPR 2025Enhancing Facial Privacy Protection via Weakening Diffusion Purification
• NeurIPS 2024Adaptive Randomized Smoothing: Certified Adversarial Robustness for Multi-Step Defences
• ICML 2024Privacy Preserving Adaptive Experiment Design
• ICLR 2024Beyond Memorization: Violating Privacy via Inference with Large Language Models
• CVPR 2024Validating Privacy-Preserving Face Recognition under a Minimum Assumption
• NeurIPS 2023Faster approximate subgraph counts with privacy
• ICML 2023GuardHFL: Privacy Guardian for Heterogeneous Federated Learning
• ICLR 2023Decepticons: Corrupted Transformers Breach Privacy in Federated Learning for Language Models
• CVPR 2023Privacy-Preserving Representations Are Not Enough: Recovering Scene Content From Camera Poses
• NeurIPS 2026Asymmetric Invertible Threat: Learning Reversible Privacy Defense for Face Recognition
• ICML 2026Accurate, Private, Secure, Federated U-statistics with Higher Degree
• ICLR 2026Measuring Physical-World Privacy Awareness of Large Language Models: An Evaluation Benchmark
• CVPR 2026LDP-Slicing Local Differential Privacy for Images via Randomized Bit-Plane Slicing

All Social aspects · privacy papers

181 papers, page 2 of 4

51. ICML 2026Multilingual Unlearning in LLMs: Transfer, Dynamics, and Reversibility
52. ICML 2026MultiPriv: Benchmarking Individual-Level Privacy Reasoning in Vision-Language Models
53. ICML 2026Near-Optimal Private Linear Regression via Iterative Hessian Mixing
54. ICML 2026OptiFluence: Principled Design of Privacy Canaries
55. ICML 2026Optimal conversion from Rényi Differential Privacy to $f$-Differential Privacy
56. ICML 2026OSNIP: Balancing the Privacy-Utility-Efficiency Trilemma in LLM Inference via Obfuscated Semantic Null Space
57. ICML 2026PEARL: Differentially Private and Entropy-Aware Regulated Language Generation
58. ICML 2026Persuasive Privacy
59. ICML 2026Preference-Calibrated Optimization with Score-Level Distribution Alignment for Text-to-Image Diffusion Model Unlearning
60. ICML 2026PRISM: Gauge-Invariant Tangent-Space Differentially Private LoRA
61. ICML 2026Privacy Amplification in Differentially Private Zeroth-Order Optimization with Hidden States
62. ICML 2026Privacy-Aware Data Integration for Enhanced Quantile Inference under Heterogeneity
63. ICML 2026Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch
64. ICML 2026Private Learning with Public Feature Conditioning
65. ICML 2026Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Data
66. ICML 2026PrivCode++: Latent-Conditioned Differentially Private Code Generation for Comprehensive Guarantees
67. ICML 2026PrivGate: Steering Contextual Integrity in LLMs via Latent Space Geometry
68. ICML 2026Provably Protecting Fine-Tuned LLMs from Training Data Extraction while Preserving Utility
69. ICML 2026RECOVER: Reliable Detection of Unauthorized Data Usage in Text-to-Image Diffusion Models via Inversion Robustness
70. ICML 2026Reverse-Engineering Model Editing on Language Models
71. ICML 2026Revisiting Asymmetries in Black-box Link Stealing against Graph Neural Networks
72. ICML 2026Revisiting ML Training under Fully Homomorphic Encryption: Convergence Guarantees, Differential Privacy, and Efficient Algorithms
73. ICML 2026Selective Disclosure Watermarking for Large Language Models
74. ICML 2026Shift-Dependent Asymmetry: Orthogonal Inverse Low-Rank Adaptation for Federated Medical Segmentation
75. ICML 2026Shuffling-Aware Optimization for Private Vector Mean Estimation
76. ICML 2026SlaClip: Gradient Norm Slacks can be Indicator for Adaptive Clipping in DP-SGD
77. ICML 2026Social Hippocampus Memory Learning
78. ICML 2026Solving Positive Linear Programs with Differential Privacy
79. ICML 2026Tackling Fake Forgetting through Uncertainty Quantification
80. ICML 2026The Double-Edged Nature of the Rashomon Set for Trustworthy Machine Learning
81. ICML 2026The Forgetting-Retention Dilemma: Certified Unlearning Theory in Continual Learning
82. ICML 2026The Hidden Risk: Membership Inference Attacks on Multimodal Federated Learning via Modality Imbalance
83. ICML 2026Towards Whole-corpus Reconstruction of Heterogeneous RAG Knowledge Bases
84. ICML 2026Trajectory-Aware Certified Decentralized Unlearning via SGD Stability
85. ICML 2026Unlearning Isn’t Forgetting: Revealing Hidden Leakage in Class Unlearning Evaluations
86. ICML 2026Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
87. ICML 2026Unlearning’s Blind Spots: Over‑Unlearning and Prototypical Relearning Attack
88. ICML 2026Variance-Reduced $(\varepsilon, \delta)-$Unlearning using Forget Set Gradients
89. ICML 2026VPD-100K: Towards Generalizable and Fine-grained Visual Privacy Protection
90. ICML 2026Who’s in Charge? Disempowerment Patterns in Real-World LLM Usage
91. ICML 2025A Certified Unlearning Approach without Access to Source Data
92. ICML 2025A Unified Theoretical Analysis of Private and Robust Offline Alignment: from RLHF to DPO
93. ICML 2025Adapting to Linear Separable Subsets with Large-Margin in Differentially Private Learning
94. ICML 2025An Efficient Private GPT Never Autoregressively Decodes
95. ICML 2025Approximate Differential Privacy of the $\ell_2$ Mechanism
96. ICML 2025Auditing $f$-differential privacy in one run
97. ICML 2025Auditing Prompt Caching in Language Model APIs
98. ICML 2025Breaking the $n^{1.5}$ Additive Error Barrier for Private and Efficient Graph Sparsification via Private Expander Decomposition
99. ICML 2025Breaking the Quadratic Barrier: Robust Cardinality Sketches for Adaptive Queries
100. ICML 2025CAN: Leveraging Clients As Navigators for Generative Replay in Federated Continual Learning

FAQ

How many Social aspects · privacy papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

431 in total: 21 in 2023, 117 in 2024, 155 in 2025, 138 in 2026.

How many Social aspects · privacy papers are at NeurIPS 2026?

NeurIPS 2026 has 44 Social aspects · privacy papers, versus 80 the year before.

How many Social aspects · privacy papers are at ICLR 2026?

ICLR 2026 has 2 Social aspects · privacy papers, versus 2 the year before.

How many Social aspects · privacy papers are at ICML 2026?

ICML 2026 has 90 Social aspects · privacy papers, versus 71 the year before.

How many Social aspects · privacy papers are at CVPR 2026?

CVPR 2026 has 2 Social aspects · privacy papers, versus 2 the year before.

Which conference publishes the most Social aspects · privacy papers?

NeurIPS, with 215 papers from 2023 to 2026.