kernel methods in the space of (short, propositional) programs!!
why memorize and interpolate answers when you can memorize and interpolate answer-producing procedures??
The approach to reasoning LLMs use looks unlike retrieval, and more like a generalisable strategy synthesising procedural knowledge from many documents doing a similar form of reasoning.
kernel methods in the space of (short, propositional) programs!!
why memorize and interpolate answers when you can memorize and interpolate answer-producing procedures??