{"total":593,"rows":[{"id":"2304.01373","title":"Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling","created":"2023-04-03","cited_by_count":1786,"area":null},{"id":"2102.08981","title":"Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Concepts","created":"2021-02-17","cited_by_count":1216,"area":null},{"id":"2112.04359","title":"Ethical and social risks of harm from Language Models","created":"2021-12-08","cited_by_count":966,"area":null},{"id":"2202.07646","title":"Quantifying Memorization Across Neural Language Models","created":"2022-02-15","cited_by_count":763,"area":null},{"id":"2211.08411","title":"Large Language Models Struggle to Learn Long-Tail Knowledge","created":"2022-11-15","cited_by_count":458,"area":null},{"id":"2110.05679","title":"Large Language Models Can Be Strong Differentially Private Learners","created":"2021-10-12","cited_by_count":412,"area":null},{"id":"2110.06500","title":"Differentially Private Fine-tuning of Language Models","created":"2021-10-13","cited_by_count":400,"area":null},{"id":"2202.06539","title":"Deduplicating Training Data Mitigates Privacy Risks in Language Models","created":"2022-02-14","cited_by_count":314,"area":null},{"id":"2403.10131","title":"RAFT: Adapting Language Model to Domain Specific RAG","created":"2024-03-15","cited_by_count":262,"area":null},{"id":"2203.08242","title":"Data Contamination: From Memorization to Exploitation","created":"2022-03-15","cited_by_count":236,"area":null},{"id":"2310.17623","title":"Proving Test Set Contamination in Black Box Language Models","created":"2023-10-26","cited_by_count":230,"area":null},{"id":"2205.10770","title":"Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Models","created":"2022-05-22","cited_by_count":220,"area":null},{"id":"2101.04535","title":"Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning","created":"2021-01-11","cited_by_count":199,"area":null},{"id":"2202.05520","title":"What Does it Mean for a Language Model to Preserve Privacy?","created":"2022-02-11","cited_by_count":195,"area":null},{"id":"2310.07298","title":"Beyond Memorization: Violating Privacy Via Inference with Large Language Models","created":"2023-10-11","cited_by_count":175,"area":null},{"id":"2309.17410","title":"Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attacks","created":"2023-09-29","cited_by_count":158,"area":null},{"id":"2112.12938","title":"Counterfactual Memorization in Neural Language Models","created":"2021-12-24","cited_by_count":156,"area":null},{"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":"2201.04845","title":"Reconstructing Training Data with Informed Adversaries","created":"2022-01-13","cited_by_count":148,"area":null},{"id":"2303.03915","title":"The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset","created":"2023-03-07","cited_by_count":148,"area":null},{"id":"2304.11158","title":"Emergent and Predictable Memorization in Large Language Models","created":"2023-04-21","cited_by_count":148,"area":null},{"id":"2302.08500","title":"Auditing large language models: a three-layered approach","created":"2023-02-16","cited_by_count":137,"area":null},{"id":"2206.10469","title":"The Privacy Onion Effect: Memorization is Relative","created":"2022-06-21","cited_by_count":136,"area":null},{"id":"2206.07758","title":"Reconstructing Training Data from Trained Neural Networks","created":"2022-06-15","cited_by_count":133,"area":null},{"id":"2102.06387","title":"The Distributed Discrete Gaussian Mechanism for Federated Learning with Secure Aggregation","created":"2021-02-12","cited_by_count":128,"area":null},{"id":"2111.09509","title":"How much do language models copy from their training data? Evaluating linguistic novelty in text generation using RAVEN","created":"2021-11-18","cited_by_count":124,"area":null},{"id":"2311.04897","title":"Future Lens: Anticipating Subsequent Tokens from a Single Hidden State","created":"2023-11-08","cited_by_count":121,"area":null},{"id":"2108.01624","title":"Large-Scale Differentially Private BERT","created":"2021-08-03","cited_by_count":118,"area":null},{"id":"2207.00099","title":"Measuring Forgetting of Memorized Training Examples","created":"2022-06-30","cited_by_count":118,"area":null},{"id":"2303.15715","title":"Foundation Models and Fair Use","created":"2023-03-28","cited_by_count":116,"area":null},{"id":"2207.00220","title":"Pile of Law: Learning Responsible Data Filtering from the Law and a 256GB Open-Source Legal Dataset","created":"2022-07-01","cited_by_count":113,"area":null},{"id":"2210.14348","title":"Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipe","created":"2022-10-25","cited_by_count":108,"area":null},{"id":"2407.12043","title":"The Art of Saying No: Contextual Noncompliance in Language Models","created":"2024-07-02","cited_by_count":105,"area":null},{"id":"2204.00032","title":"Truth Serum: Poisoning Machine Learning Models to Reveal Their Secrets","created":"2022-03-31","cited_by_count":101,"area":null},{"id":"2210.17546","title":"Preventing Verbatim Memorization in Language Models Gives a False Sense of Privacy","created":"2022-10-31","cited_by_count":100,"area":null},{"id":"2302.08091","title":"Do We Still Need Clinical Language Models?","created":"2023-02-16","cited_by_count":98,"area":null},{"id":"2407.11016","title":"LongLaMP: A Benchmark for Personalized Long-form Text Generation","created":"2024-06-27","cited_by_count":97,"area":null},{"id":"2107.06641","title":"Trustworthy AI: A Computational Perspective","created":"2021-07-12","cited_by_count":93,"area":null},{"id":"2404.14851","title":"From Matching to Generation: A Survey on Generative Information Retrieval","created":"2024-04-23","cited_by_count":89,"area":null},{"id":"2106.09352","title":"Large Scale Private Learning via Low-rank Reparametrization","created":"2021-06-17","cited_by_count":88,"area":null},{"id":"2212.06470","title":"Position: Considerations for Differentially Private Learning with Large-Scale Public Pretraining","created":"2022-12-13","cited_by_count":88,"area":null},{"id":"2201.12328","title":"Toward Training at ImageNet Scale with Differential Privacy","created":"2022-01-28","cited_by_count":87,"area":null},{"id":"2102.12677","title":"Do Not Let Privacy Overbill Utility: Gradient Embedding Perturbation for Private Learning","created":"2021-02-25","cited_by_count":86,"area":null},{"id":"2111.00197","title":"Backdoor Pre-trained Models Can Transfer to All","created":"2021-10-30","cited_by_count":86,"area":null},{"id":"2102.05429","title":"Node-Level Membership Inference Attacks Against Graph Neural Networks","created":"2021-02-10","cited_by_count":84,"area":null},{"id":"2103.06191","title":"A Study of Face Obfuscation in ImageNet","created":"2021-03-10","cited_by_count":83,"area":null},{"id":"2101.06570","title":"Membership Inference Attack on Graph Neural Networks","created":"2021-01-17","cited_by_count":81,"area":null},{"id":"2305.03010","title":"Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence","created":"2023-05-04","cited_by_count":81,"area":null},{"id":"2201.06820","title":"Recommendation Unlearning","created":"2022-01-18","cited_by_count":80,"area":null},{"id":"2310.20138","title":"DEPN: Detecting and Editing Privacy Neurons in Pretrained Language Models","created":"2023-10-31","cited_by_count":79,"area":null},{"id":"2107.03451","title":"Anticipating Safety Issues in E2E Conversational AI: Framework and Tooling","created":"2021-07-07","cited_by_count":78,"area":null},{"id":"2203.07618","title":"Do Language Models Plagiarize?","created":"2022-03-15","cited_by_count":73,"area":null},{"id":"2301.13105","title":"Generalization on the Unseen, Logic Reasoning and Degree Curriculum","created":"2023-01-30","cited_by_count":73,"area":null},{"id":"2205.08514","title":"Recovering Private Text in Federated Learning of Language Models","created":"2022-05-17","cited_by_count":70,"area":null},{"id":"2302.04460","title":"Bag of Tricks for Training Data Extraction from Language Models","created":"2023-02-09","cited_by_count":70,"area":null},{"id":"2305.18465","title":"Federated Learning of Gboard Language Models with Differential Privacy","created":"2023-05-29","cited_by_count":66,"area":null},{"id":"2210.01296","title":"Recitation-Augmented Language Models","created":"2022-10-04","cited_by_count":65,"area":null},{"id":"2305.15560","title":"Differentially Private Synthetic Data via Foundation Model APIs 1: Images","created":"2023-05-24","cited_by_count":65,"area":null},{"id":"2306.01684","title":"Harnessing large-language models to generate private synthetic text","created":"2023-06-02","cited_by_count":65,"area":null},{"id":"2310.16111","title":"Locally Differentially Private Document Generation Using Zero Shot Prompting","created":"2023-10-24","cited_by_count":63,"area":null},{"id":"2108.12944","title":"Selective Differential Privacy for Language Modeling","created":"2021-08-30","cited_by_count":62,"area":null},{"id":"2210.03588","title":"Understanding Transformer Memorization Recall Through Idioms","created":"2022-10-07","cited_by_count":62,"area":null},{"id":"2505.24832","title":"How much do language models memorize?","created":"2025-05-30","cited_by_count":61,"area":null},{"id":"2101.11073","title":"Property Inference From Poisoning","created":"2021-01-26","cited_by_count":60,"area":null},{"id":"2104.14222","title":"Privacy-Preserving Portrait Matting","created":"2021-04-29","cited_by_count":60,"area":null},{"id":"2106.03216","title":"On Memorization in Probabilistic Deep Generative Models","created":"2021-06-06","cited_by_count":60,"area":null},{"id":"2301.02344","title":"TrojanPuzzle: Covertly Poisoning Code-Suggestion Models","created":"2023-01-06","cited_by_count":60,"area":null},{"id":"2202.08312","title":"Improved Differential Privacy for SGD via Optimal Private Linear Operators on Adaptive Streams","created":"2022-02-16","cited_by_count":59,"area":null},{"id":"2111.11343","title":"Decentralized Federated Learning through Proxy Model Sharing","created":"2021-11-22","cited_by_count":57,"area":null},{"id":"2407.17817","title":"Demystifying Verbatim Memorization in Large Language Models","created":"2024-07-25","cited_by_count":56,"area":null},{"id":"2505.15146","title":"lmgame-Bench: How Good are LLMs at Playing Games?","created":"2025-05-21","cited_by_count":55,"area":null},{"id":"2203.11481","title":"Mixed Differential Privacy in Computer Vision","created":"2022-03-22","cited_by_count":54,"area":null},{"id":"2402.00786","title":"CroissantLLM: A Truly Bilingual French-English Language Model","created":"2024-02-01","cited_by_count":53,"area":null},{"id":"2106.03408","title":"Antipodes of Label Differential Privacy: PATE and ALIBI","created":"2021-06-07","cited_by_count":52,"area":null},{"id":"2207.01193","title":"A Customized Text Sanitization Mechanism with Differential Privacy","created":"2022-07-04","cited_by_count":52,"area":null},{"id":"2210.03403","title":"TAN Without a Burn: Scaling Laws of DP-SGD","created":"2022-10-07","cited_by_count":52,"area":null},{"id":"2212.01539","title":"Exploring the Limits of Differentially Private Deep Learning with Group-wise Clipping","created":"2022-12-03","cited_by_count":52,"area":null},{"id":"2212.10986","title":"SoK: Let the Privacy Games Begin! A Unified Treatment of Data Inference Privacy in Machine Learning","created":"2022-12-21","cited_by_count":51,"area":null},{"id":"2406.10209","title":"Be like a Goldfish, Don't Memorize! Mitigating Memorization in Generative LLMs","created":"2024-06-14","cited_by_count":50,"area":null},{"id":"2209.01292","title":"Are Attribute Inference Attacks Just Imputation?","created":"2022-09-02","cited_by_count":49,"area":null},{"id":"2106.02990","title":"Self-Damaging Contrastive Learning","created":"2021-06-06","cited_by_count":48,"area":null},{"id":"2109.11429","title":"Robin Hood and Matthew Effects: Differential Privacy Has Disparate Impact on Synthetic Data","created":"2021-09-23","cited_by_count":48,"area":null},{"id":"2211.06530","title":"Multi-Epoch Matrix Factorization Mechanisms for Private Machine Learning","created":"2022-11-12","cited_by_count":48,"area":null},{"id":"2106.13329","title":"Covariance-Aware Private Mean Estimation Without Private Covariance Estimation","created":"2021-06-24","cited_by_count":46,"area":null},{"id":"2201.12383","title":"Bounding Training Data Reconstruction in Private (Deep) Learning","created":"2022-01-28","cited_by_count":46,"area":null},{"id":"2205.12506","title":"Memorization in NLP Fine-tuning Methods","created":"2022-05-25","cited_by_count":45,"area":null},{"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":"2105.01622","title":"Poisoning the Unlabeled Dataset of Semi-Supervised Learning","created":"2021-05-04","cited_by_count":44,"area":null},{"id":"2107.01614","title":"Survey: Leakage and Privacy at Inference Time","created":"2021-07-04","cited_by_count":44,"area":null},{"id":"2111.15521","title":"Node-Level Differentially Private Graph Neural Networks","created":"2021-11-23","cited_by_count":43,"area":null},{"id":"2406.18664","title":"Evaluating Copyright Takedown Methods for Language Models","created":"2024-06-26","cited_by_count":43,"area":null},{"id":"2205.02973","title":"Large Scale Transfer Learning for Differentially Private Image Classification","created":"2022-05-06","cited_by_count":41,"area":null},{"id":"2312.17493","title":"Differentially Private Low-Rank Adaptation of Large Language Model Using Federated Learning","created":"2023-12-29","cited_by_count":41,"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":"2410.19482","title":"Measuring memorization in language models via probabilistic extraction","created":"2024-10-25","cited_by_count":41,"area":null},{"id":"2203.03550","title":"When BERT Meets Quantum Temporal Convolution Learning for Text Classification in Heterogeneous Computing","created":"2022-02-17","cited_by_count":40,"area":null},{"id":"2305.11759","title":"Controlling the Extraction of Memorized Data from Large Language Models via Prompt-Tuning","created":"2023-05-19","cited_by_count":40,"area":null},{"id":"2309.05610","title":"Privacy Side Channels in Machine Learning Systems","created":"2023-09-11","cited_by_count":40,"area":null},{"id":"2310.10628","title":"Data Contamination Through the Lens of Time","created":"2023-10-16","cited_by_count":40,"area":null},{"id":"2203.03761","title":"The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning","created":"2022-03-07","cited_by_count":39,"area":null}],"next_cursor":"100"}