Apparently human grandmasters remember around 50,000 board positions and play moves that manipulate the current board position to a more favourable one. They d

Apparently human grandmasters remember around 50,000 board positions and play moves that manipulate the current board position to a more favourable one. They don't see moves in the same fashion that a beginner does. It seems like the LLMs are closer to the approach used by grandmasters.