Nathan Lambert· @natolambert · X·· 1 天前AI 评分41
AI 导读
当前的 harness 在 LLM 如何管理自身上下文方面设置非常僵化。我们展示了你可以让 LLM 直接编辑上下文——从而实现更好的长时程记忆、FLOP 高效适配，以及更多。 不用再害怕 compaction 了！
正文
Current harnesses have a very rigid setup in how the LLM can manage its own context. We show that you can let LLMs edit the context directly - enabling better long-horizon memory, FLOP efficient adaption, & more.
Don't fear compaction anymore!
‼️The Bitter Lesson for context management: Giving LMs unrestricted control over their context beats human-designed SOTA! Introducing 🩵Context Language Models (CLMs)🩵 - Natively manage their own context - Treat context as a file - Learn policies in CLM weights, no harness在 X 查看被引用的帖子
来源：Nathan Lambert · x.com