Google Research proposes a sleep framework to help LLMs retain knowledge
AIGoogle Research describes a sleep framework that alternates active interaction with offline memory consolidation so LLMs can keep new knowledge without catastrophic forgetting. The post reports results on unseen-language translation, long-context retrieval up to 10M tokens, factual knowledge incorporation, and AIME 24/25 and HMMT 25 math benchmarks, where the sleep-enhanced base model outperformed SFT, GRPO, and OPSD baselines.
AIWhy it matters.
The post explains a sleep-style consolidation method for continual learning, with benchmark results against fine-tuning and in-context baselines that show whether it avoids catastrophic forgetting.
Loading source posts…
Source details
Google Research · research.googlePublished · added here
Original title: The sleep framework: A new way for LLMs to retain knowledge in continual learning