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.
Machine unlearning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
CONDITIONAL 2representative citing papers
TOFU is a new benchmark with synthetic profiles and metrics demonstrating that existing unlearning algorithms for LLMs fail to achieve effective forgetting of targeted information.
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.
-
TOFU: A Task of Fictitious Unlearning for LLMs
TOFU is a new benchmark with synthetic profiles and metrics demonstrating that existing unlearning algorithms for LLMs fail to achieve effective forgetting of targeted information.