Fetching the paper…

Split, Unlearn, Merge: Leveraging Data Attributes for More Effective Unlearning in LLMs · Around