作者位次 #6/7ACL 2026Findings· Findings of ACL
Understanding Secret Leakage Risks in Code LLM s: A Tokenization Perspective
Meifang Chen · Zhe Yang · Huang Nianchen · Yizhan Huang · Yichen LI · Zihan Li · Michael R. Lyu
摘要
Code secrets are sensitive assets for software developers, and their leakage poses significant cybersecurity risks. While the rapid development of AI code assistants powered by Code Large Language Models (CLLMs), CLLMs are shown to inadvertently leak such secrets due to a notorious memorization p...