Memories Retrieved from Many Paths: A Multi-Prefix Framework for Robust Detection of Training Data Leakage in Large Language Models

Documenti analoghi

• Topology Matters: Measuring Memory Leakage in Multi-Agent LLMs
  di: Liu, Jinbo, et al.
  Pubblicazione: (2025)
• Prefix Guidance: A Steering Wheel for Large Language Models to Defend Against Jailbreak Attacks
  di: Zhao, Jiawei, et al.
  Pubblicazione: (2024)
• Quantifying Association Capabilities of Large Language Models and Its Implications on Privacy Leakage
  di: Shao, Hanyin, et al.
  Pubblicazione: (2023)
• Token-wise Influential Training Data Retrieval for Large Language Models
  di: Lin, Huawei, et al.
  Pubblicazione: (2024)
• Prefix Probing: Lightweight Harmful Content Detection for Large Language Models
  di: Yang, Jirui, et al.
  Pubblicazione: (2025)