Min-K% Prob detects pretraining data in LLMs by flagging outlier low-probability words in text, achieving 7.4% better performance than prior methods on the new WIKIMIA benchmark.
Quantifying privacy risks of masked language models using membership inference attacks
6 Pith papers cite this work, alongside 10 external citations. Polarity classification is still indexing.
6
Pith papers citing it
10
external citations · external index
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Batch normalization amplifies memorization of outlier samples in deep neural networks, directly increasing susceptibility to membership inference attacks.
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
LoRA-MINT uses perplexity to perform membership inference on LoRA-fine-tuned LLMs, reporting 0.77-0.92 precision across four models and three datasets while outperforming baselines.
Authors introduce MLM and CLM specialization methods that avoid memorizing identifiers in sensitive training data while aiming for a privacy-utility tradeoff on medical datasets.
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.
citing papers explorer
-
Detecting Pretraining Data from Large Language Models
Min-K% Prob detects pretraining data in LLMs by flagging outlier low-probability words in text, achieving 7.4% better performance than prior methods on the new WIKIMIA benchmark.
-
Batch Normalization Amplifies Memorization and Privacy Risks
Batch normalization amplifies memorization of outlier samples in deep neural networks, directly increasing susceptibility to membership inference attacks.
-
Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications
First unified survey formalizing Pretraining Data Exposure across exposure levels and reviewing attack, defense, and contamination methods for LLMs.
-
Auditing Training Data in Domain-adapted LLMs: LoRA-MINT
LoRA-MINT uses perplexity to perform membership inference on LoRA-fine-tuned LLMs, reporting 0.77-0.92 precision across four models and three datasets while outperforming baselines.
-
Towards the Anonymization of the Language Modeling
Authors introduce MLM and CLM specialization methods that avoid memorizing identifiers in sensitive training data while aiming for a privacy-utility tradeoff on medical datasets.
-
Large Language Model Agent: A Survey on Methodology, Applications and Challenges
A survey that deconstructs LLM agent systems via a methodology-centered taxonomy linking design principles to emergent behaviors, applications, and challenges.