On: Ladder: Self-improving LLMs through recursive problem decomposition

[esafak]

The "massive" context is still tiny compared with the size of the model whose weights you propose to update.