Making AI models 'forget' undesirable data hurts their performance | TechCrunchCurrent unlearning techniques for AI models can lead to degradation and reduced capabilities, impacting their usability in real-world scenarios.
Who's Harry Potter? Approximate Unlearning in LLMs: Description of our technique | HackerNoonUnlearning specific content from a generative language model by negating the loss function does not yield promising results and can lead to unintended consequences.