Don't Forget It! Conditional Sparse Autoencoder Clamping Works for Unlearning

Documenti analoghi

• Controlling for Unobserved Confounding with Large Language Model Classification of Patient Smoking Status
  di: Lee, Samuel, et al.
  Pubblicazione: (2024)
• Don't Forget Imagination!
  di: Vityaev, Evgenii E., et al.
  Pubblicazione: (2025)
• "I've Seen How This Goes": Characterizing Diversity via Progressive Conditional Surprise
  di: Khoriaty, Matthew, et al.
  Pubblicazione: (2026)
• F-GRPO: Don't Let Your Policy Learn the Obvious and Forget the Rare
  di: Plyusov, Daniil, et al.
  Pubblicazione: (2026)
• Kolmogorov Complexity Bounds for LLM Steganography and a Perplexity-Based Detection Proxy
  di: Shportko, Andrii
  Pubblicazione: (2026)