Identifying Pre-training Data in LLMs: A Neuron Activation-Based Detection Framework

Documenti analoghi

• DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction
  di: Tang, Hongyi, et al.
  Pubblicazione: (2026)
• Probing Language Models for Pre-training Data Detection
  di: Liu, Zhenhua, et al.
  Pubblicazione: (2024)
• A Pre-training Framework for Relational Data with Information-theoretic Principles
  di: Truong, Quang, et al.
  Pubblicazione: (2025)
• Automated Detection of Pre-training Text in Black-box LLMs
  di: Hu, Ruihan, et al.
  Pubblicazione: (2025)
• Exploring the Benefit of Activation Sparsity in Pre-training
  di: Zhang, Zhengyan, et al.
  Pubblicazione: (2024)