{"total":88,"rows":[{"id":"2301.13188","title":"Extracting Training Data from Diffusion Models","created":"2023-01-30","cited_by_count":732,"area":"diffusion"},{"id":"2305.18462","title":"Membership Inference Attacks against Language Models via Neighbourhood Comparison","created":"2023-05-29","cited_by_count":258,"area":"safety"},{"id":"2402.00888","title":"Security and Privacy Challenges of Large Language Models: A Survey","created":"2024-01-30","cited_by_count":186,"area":"safety"},{"id":"2303.00654","title":"How to DP-fy ML: A Practical Guide to Machine Learning with Differential Privacy","created":"2023-03-01","cited_by_count":153,"area":null},{"id":"2403.03329","title":"Guardrail Baselines for Unlearning in LLMs","created":"2024-03-05","cited_by_count":96,"area":"safety"},{"id":"2302.13861","title":"Differentially Private Diffusion Models Generate Useful Synthetic Images","created":"2023-02-27","cited_by_count":86,"area":"diffusion"},{"id":"2403.01749","title":"Differentially Private Synthetic Data via Foundation Model APIs 2: Text","created":"2024-03-04","cited_by_count":76,"area":"synth"},{"id":"2410.02879","title":"Position: LLM Unlearning Benchmarks are Weak Measures of Progress","created":"2024-10-03","cited_by_count":63,"area":"safety"},{"id":"2305.16157","title":"Training Data Extraction From Pre-trained Language Models: A Survey","created":"2023-05-25","cited_by_count":45,"area":null},{"id":"2408.12787","title":"LLM-PBE: Assessing Data Privacy in Large Language Models","created":"2024-08-23","cited_by_count":41,"area":null},{"id":"2302.09483","title":"Why Is Public Pretraining Necessary for Private Model Training?","created":"2023-02-19","cited_by_count":37,"area":null},{"id":"2402.13659","title":"Privacy-Preserving Instructions for Aligning Large Language Models","created":"2024-02-21","cited_by_count":35,"area":"alignment"},{"id":"2405.10853","title":"The Future of Large Language Model Pre-training is Federated","created":"2024-05-17","cited_by_count":30,"area":null},{"id":"2305.15759","title":"DP-LDMs: Differentially Private Latent Diffusion Models","created":"2023-05-25","cited_by_count":29,"area":"diffusion"},{"id":"2406.02958","title":"PrE-Text: Training Language Models on Private Federated Data in the Age of LLMs","created":"2024-06-05","cited_by_count":29,"area":null},{"id":"2305.13865","title":"Selective Pre-training for Private Fine-tuning","created":"2023-05-23","cited_by_count":23,"area":null},{"id":"2306.06076","title":"Differentially Private Image Classification by Learning Priors from Random Processes","created":"2023-06-08","cited_by_count":19,"area":null},{"id":"2412.18295","title":"Pirates of the RAG: Adaptively Attacking LLMs to Leak Knowledge Bases","created":"2024-12-24","cited_by_count":18,"area":"rag"},{"id":"2303.12247","title":"Exploring the Benefits of Visual Prompting in Differential Privacy","created":"2023-03-22","cited_by_count":17,"area":null},{"id":"2404.04360","title":"Prompt Public Large Language Models to Synthesize Data for Private On-device Applications","created":"2024-04-05","cited_by_count":17,"area":null},{"id":"2502.21278","title":"Does Generation Require Memorization? Creative Diffusion Models using Ambient Diffusion","created":"2025-02-28","cited_by_count":17,"area":"diffusion"},{"id":"2303.01256","title":"Choosing Public Datasets for Private Machine Learning via Gradient Subspace Distance","created":"2023-03-02","cited_by_count":16,"area":null},{"id":"2308.10888","title":"Unlocking Accuracy and Fairness in Differentially Private Image Classification","created":"2023-08-21","cited_by_count":16,"area":null},{"id":"2311.15792","title":"Rethinking Privacy in Machine Learning Pipelines from an Information Flow Control Perspective","created":"2023-11-27","cited_by_count":16,"area":null},{"id":"2510.15001","title":"VaultGemma: A Differentially Private Gemma Model","created":"2025-10-15","cited_by_count":16,"area":null},{"id":"2502.19726","title":"Tokens for Learning, Tokens for Unlearning: Mitigating Membership Inference Attacks in Large Language Models via Dual-Purpose Training","created":"2025-02-27","cited_by_count":15,"area":"safety"},{"id":"2306.01181","title":"TMI! Finetuned Models Leak Private Information from their Pretraining Data","created":"2023-06-01","cited_by_count":14,"area":null},{"id":"2308.06239","title":"Private Distribution Learning with Public Data: The View from Sample Compression","created":"2023-08-11","cited_by_count":14,"area":null},{"id":"2311.12850","title":"PrivImage: Differentially Private Synthetic Image Generation using Diffusion Models with Semantic-Aware Pretraining","created":"2023-10-19","cited_by_count":13,"area":"diffusion"},{"id":"2302.01190","title":"On the Efficacy of Differentially Private Few-shot Image Classification","created":"2023-02-02","cited_by_count":12,"area":null},{"id":"2302.09042","title":"Privately Customizing Prefinetuning to Better Match User Data in Federated Learning","created":"2023-02-17","cited_by_count":12,"area":null},{"id":"2309.16398","title":"Recent Advances of Differential Privacy in Centralized Deep Learning: A Systematic Survey","created":"2023-09-28","cited_by_count":12,"area":null},{"id":"2512.10169","title":"The 2025 Foundation Model Transparency Index","created":"2025-12-11","cited_by_count":11,"area":null},{"id":"2306.03962","title":"PILLAR: How to make semi-private learning more effective","created":"2023-06-06","cited_by_count":10,"area":null},{"id":"2507.02974","title":"InvisibleInk: High-Utility and Low-Cost Text Generation with Differential Privacy","created":"2025-06-30","cited_by_count":10,"area":"rag"},{"id":"2402.18752","title":"Pre-training Differentially Private Models with Limited Public Data","created":"2024-02-28","cited_by_count":8,"area":null},{"id":"2306.08173","title":"Safeguarding Data in Multimodal AI: A Differentially Private Approach to CLIP Training","created":"2023-06-13","cited_by_count":7,"area":null},{"id":"2311.11369","title":"Optimal Locally Private Nonparametric Classification with Public Data","created":"2023-11-19","cited_by_count":7,"area":null},{"id":"2502.05087","title":"Mitigating Unintended Memorization with LoRA in Federated Learning for LLMs","created":"2025-02-07","cited_by_count":7,"area":null},{"id":"2506.19882","title":"Position: Machine Learning Conferences Should Establish a \"Refutations and Critiques\" Track","created":"2025-06-24","cited_by_count":7,"area":null},{"id":"2410.03026","title":"Estimating Privacy Leakage of Augmented Contextual Knowledge in Language Models","created":"2024-10-03","cited_by_count":6,"area":null},{"id":"2506.05126","title":"Membership Inference Attacks on Sequence Models","created":"2025-06-05","cited_by_count":6,"area":"safety"},{"id":"2507.01129","title":"On Design Principles for Private Adaptive Optimizers","created":"2025-07-01","cited_by_count":6,"area":null},{"id":"2312.15551","title":"On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift","created":"2023-12-24","cited_by_count":5,"area":null},{"id":"2405.18194","title":"Delving into Differentially Private Transformer","created":"2024-05-28","cited_by_count":5,"area":null},{"id":"2502.06555","title":"Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?","created":"2025-02-10","cited_by_count":5,"area":null},{"id":"2301.13104","title":"Equivariant Differentially Private Deep Learning: Why DP-SGD Needs Sparser Models","created":"2023-01-30","cited_by_count":4,"area":null},{"id":"2309.15696","title":"A Unified View of Differentially Private Deep Generative Modeling","created":"2023-09-27","cited_by_count":4,"area":null},{"id":"2503.20846","title":"Generating Synthetic Data with Formal Privacy Guarantees: State of the Art and the Road Ahead","created":"2025-03-26","cited_by_count":4,"area":null},{"id":"2504.17421","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","created":"2025-04-24","cited_by_count":4,"area":null},{"id":"2505.05707","title":"Crowding Out The Noise: Algorithmic Collective Action Under Differential Privacy","created":"2025-05-09","cited_by_count":4,"area":null},{"id":"2509.14594","title":"SynBench: A Benchmark for Differentially Private Text Generation","created":"2025-09-18","cited_by_count":4,"area":"safety"},{"id":"2606.09401","title":"Benchmarking Empirical Privacy Protection for Adaptations of Large Language Models","created":"2026-06-08","cited_by_count":4,"area":null},{"id":"2410.17566","title":"Differentially Private Learning Needs Better Model Initialization and Self-Distillation","created":"2024-10-23","cited_by_count":3,"area":"distill"},{"id":"2412.21123","title":"ExpShield: Safeguarding Web Text from Unauthorized Crawling and LLM Exploitation","created":"2024-12-30","cited_by_count":3,"area":"safety"},{"id":"2509.22214","title":"A Law of Data Reconstruction for Random Features (and Beyond)","created":"2025-09-26","cited_by_count":3,"area":null},{"id":"2511.06305","title":"Setting $\\varepsilon$ is not the Issue in Differential Privacy","created":"2025-11-09","cited_by_count":3,"area":null},{"id":"2406.08039","title":"Differentially Private Prototypes for Imbalanced Transfer Learning","created":"2024-06-12","cited_by_count":2,"area":null},{"id":"2406.08569","title":"Noise-Aware Differentially Private Regression via Meta-Learning","created":"2024-06-12","cited_by_count":2,"area":null},{"id":"2411.03730","title":"NeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQA","created":"2024-11-06","cited_by_count":2,"area":null},{"id":"2502.06374","title":"Hyperparameters in Score-Based Membership Inference Attacks","created":"2025-02-10","cited_by_count":2,"area":"safety"},{"id":"2601.19618","title":"The pretraining domain outweighs the training objective in setting the privacy-utility trade-off of differentially private medical image analysis","created":"2026-01-27","cited_by_count":2,"area":null},{"id":"2312.04590","title":"Reconciling AI Performance and Data Reconstruction Resilience for Medical Imaging","created":"2023-12-05","cited_by_count":1,"area":null},{"id":"2408.04313","title":"Better Locally Private Sparse Estimation Given Multiple Samples Per User","created":"2024-08-08","cited_by_count":1,"area":null},{"id":"2411.12451","title":"Empirical Privacy Evaluations of Generative and Predictive Machine Learning Models -- A review and challenges for practice","created":"2024-11-19","cited_by_count":1,"area":null},{"id":"2505.20095","title":"Spurious Privacy Leakage in Neural Networks","created":"2025-05-26","cited_by_count":1,"area":null},{"id":"2506.06488","title":"Membership Inference Attacks for Unseen Classes","created":"2025-06-06","cited_by_count":1,"area":"safety"},{"id":"2507.07947","title":"Reconstructing Template-Memorized Images from Natural Prompts","created":"2025-07-10","cited_by_count":1,"area":"diffusion"},{"id":"2604.25110","title":"Knowledge Distillation Must Account for What It Loses","created":"2026-04-28","cited_by_count":1,"area":"distill"},{"id":"2606.28479","title":"Decomposing Memorization Reduction in Privacy-Preserving Fine-Tuning of SLMs for CSIRTs","created":"2026-06-26","cited_by_count":1,"area":null},{"id":"2306.08838","title":"Differentially Private Domain Adaptation with Theoretical Guarantees","created":"2023-06-15","cited_by_count":0,"area":null},{"id":"2410.00542","title":"Differentially Private Active Learning: Balancing Effective Data Selection and Privacy","created":"2024-10-01","cited_by_count":0,"area":null},{"id":"2411.14639","title":"Differentially Private Adaptation of Diffusion Models via Noisy Aggregated Embeddings","created":"2024-11-22","cited_by_count":0,"area":"diffusion"},{"id":"2504.00952","title":"Personalized Federated Training of Diffusion Models with Privacy Guarantees","created":"2025-04-01","cited_by_count":0,"area":"diffusion"},{"id":"2507.17895","title":"Lower Bounds for Public-Private Learning under Distribution Shift","created":"2025-07-23","cited_by_count":0,"area":null},{"id":"2508.15141","title":"Towards Reliable and Generalizable Differentially Private Machine Learning (Extended Version)","created":"2025-08-21","cited_by_count":0,"area":null},{"id":"2510.05753","title":"Empirical Comparison of Membership Inference Attacks in Deep Transfer Learning","created":"2025-10-07","cited_by_count":0,"area":"safety"},{"id":"2511.19498","title":"Hierarchical Dual-Strategy Unlearning for Biomedical and Healthcare Intelligence Using Imperfect and Privacy-Sensitive Medical Data","created":"2025-11-23","cited_by_count":0,"area":"safety"},{"id":"2601.06368","title":"From Easy to Hard++: Promoting Differentially Private Image Synthesis Through Spatial-Frequency Curriculum","created":"2026-01-10","cited_by_count":0,"area":null},{"id":"2601.17189","title":"Rethinking Benchmarks for Differentially Private Image Classification","created":"2026-01-23","cited_by_count":0,"area":null},{"id":"2605.11170","title":"Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data","created":"2026-05-11","cited_by_count":0,"area":"safety"},{"id":"2605.16471","title":"From AI-Generated Content to Agentic Action: Security and Safety Threats in Generative AI","created":"2026-05-15","cited_by_count":0,"area":"agents"},{"id":"2606.01849","title":"ContinuousBench: Can Differentially Private Synthetic Text Improve Capabilities?","created":"2026-06-01","cited_by_count":0,"area":null},{"id":"2606.04375","title":"When Do Fewer Coordinates Suffice in DP-SGD?","created":"2026-06-03","cited_by_count":0,"area":null},{"id":"2608.05040","title":"Private Direct Preference Optimization for LLM Alignment","created":"2026-08-05","cited_by_count":0,"area":"alignment"},{"id":"2608.26324","title":"Privacy Without Regret: Differentially Private Inference-Time Alignment","created":"2026-08-26","cited_by_count":0,"area":"alignment"},{"id":"2609.00492","title":"The Privacy-Hallucination Tradeoff in Differentially Private Language Models","created":"2026-08-31","cited_by_count":0,"area":null},{"id":"2610.06636","title":"Differentially Private Mixing of Public Datasets Improves Private Learning","created":"2026-10-05","cited_by_count":0,"area":null}],"next_cursor":null}