作者位次 #6/7ACL 2026Poster· Natural Language Processing
PDR : A Plug-and-Play Positional Decay Framework for LLM Pre-training Data Detection
Jinhan Liu · Yibo Yang · Ruiying Lu · Piotr Piękos · Yimeng Chen · Peng Wang · Dandan Guo
摘要
Detecting pre-training data in Large Language Models (LLMs) is crucial for auditing data privacy and copyright compliance, yet it remains challenging in black-box, zero-shot settings where computational resources and training data are scarce. While existing likelihood-based methods have shown pro...