SAP: Corrective Machine Unlearning with Scaled Activation Projection for Label Noise Robustness

Documenti analoghi

• Deep Unlearning: Fast and Efficient Gradient-free Approach to Class Forgetting
  di: Kodge, Sangamesh, et al.
  Pubblicazione: (2023)
• GradientSpace: Unsupervised Data Clustering for Improved Instruction Tuning
  di: Sridharan, Shrihari, et al.
  Pubblicazione: (2025)
• Finding the Muses: Identifying Coresets through Loss Trajectories
  di: Nagaraj, Manish, et al.
  Pubblicazione: (2025)
• Eigen Attention: Attention in Low-Rank Space for KV Cache Compression
  di: Saxena, Utkarsh, et al.
  Pubblicazione: (2024)
• Unveiling Privacy, Memorization, and Input Curvature Links
  di: Ravikumar, Deepak, et al.
  Pubblicazione: (2024)