一作ACL 2026Poster· Natural Language Processing
Gap-K%: Measuring Top-1 Prediction Gap for Detecting Pretraining Data
Minseo Kwak · Jaehyung Kim
摘要
The opacity of massive pretraining corpora in Large Language Models (LLMs) raises significant privacy and copyright concerns, making pretraining data detection a critical challenge.Existing state-of-the-art methods typically rely on token likelihoods, yet they often overlook the gap between the t...