A classifier trained only on transformer fine-tuning data detects an invariant memorization signature that transfers to Mamba, RWKV-4, and RecurrentGemma with AUCs of 0.963, 0.972, and 0.936.
Do membership inference attacks work on large language models?arXiv preprint arXiv:2402.07841
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5representative citing papers
DistractMIA performs output-only black-box membership inference on vision-language models by inserting semantic distractors and measuring shifts in generated text responses.
Discriminative factorization distinguishes high-quality query sets for black-box model classification, with chance-level error decaying exponentially in query budget and parameters predicting empirical decay rates on auditing tasks.
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.
CoLA reveals that subset training creates new privacy leakage surfaces via side-channel metadata and model outputs, enabling training-membership and selection-participation membership inference attacks.
citing papers explorer
-
Learning the Signature of Memorization in Autoregressive Language Models
A classifier trained only on transformer fine-tuning data detects an invariant memorization signature that transfers to Mamba, RWKV-4, and RecurrentGemma with AUCs of 0.963, 0.972, and 0.936.
-
DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction
DistractMIA performs output-only black-box membership inference on vision-language models by inserting semantic distractors and measuring shifts in generated text responses.
-
Black-box model classification under the discriminative factorization
Discriminative factorization distinguishes high-quality query sets for black-box model classification, with chance-level error decaying exponentially in query budget and parameters predicting empirical decay rates on auditing tasks.
-
Pop Quiz Attack: Black-box Membership Inference Attacks Against Large Language Models
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.
-
CoLA: A Choice Leakage Attack Framework to Expose Privacy Risks in Subset Training
CoLA reveals that subset training creates new privacy leakage surfaces via side-channel metadata and model outputs, enabling training-membership and selection-participation membership inference attacks.