Understanding the Collapse of LLMs in Model Editing

Documenti analoghi

• The Mirage of Model Editing: Revisiting Evaluation in the Wild
  di: Yang, Wanli, et al.
  Pubblicazione: (2025)
• The Butterfly Effect of Model Editing: Few Edits Can Trigger Large Language Models Collapse
  di: Yang, Wanli, et al.
  Pubblicazione: (2024)
• GloSS over Toxicity: Understanding and Mitigating Toxicity in LLMs via Global Toxic Subspace
  di: Duan, Zenghao, et al.
  Pubblicazione: (2025)
• Related Knowledge Perturbation Matters: Rethinking Multiple Pieces of Knowledge Editing in Same-Subject
  di: Duan, Zenghao, et al.
  Pubblicazione: (2025)
• Blinded by Generated Contexts: How Language Models Merge Generated and Retrieved Contexts When Knowledge Conflicts?
  di: Tan, Hexiang, et al.
  Pubblicazione: (2024)