How Training Data Shapes the Use of Parametric and In-Context Knowledge in Language Models

Documenti analoghi

• Bilinear representation mitigates reversal curse and enables consistent model editing
  di: Kim, Dong-Kyum, et al.
  Pubblicazione: (2025)
• Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning
  di: Yang, Nakyeong, et al.
  Pubblicazione: (2025)
• FaithUn: Toward Faithful Forgetting in Language Models by Investigating the Interconnectedness of Knowledge
  di: Yang, Nakyeong, et al.
  Pubblicazione: (2025)
• Rethinking Post-Unlearning Behavior of Large Vision-Language Models
  di: Kim, Minsung, et al.
  Pubblicazione: (2025)
• MVMR: A New Framework for Evaluating Faithfulness of Video Moment Retrieval against Multiple Distractors
  di: Yang, Nakyeong, et al.
  Pubblicazione: (2023)