Unmemorization in Large Language Models via Self-Distillation and Deliberate Imagination

arXiv:2402.10052v1 Announce Type: new
Abstract: While displaying impressive generation capabilities across many tasks, Large Language Models (LLMs) still struggle with crucial issues of privacy violation and unwanted exposure of sensitive data. This raises an essential question: how should we prevent such undesired behavior of LLMs while maintaining their strong generation and natural language understanding (NLU) capabilities? In this work, we introduce a novel approach termed deliberate imagination in the context of LLM unlearning. Instead of trying to forget memorized data, …

abstract arxiv behavior capabilities cs.ai cs.cl data distillation imagination language language models large language large language models llms privacy question raises struggle tasks type via