Tag
A discussion on the sources of recent AI advancements, noting that post-training, fine-tuning, and reinforcement learning have become key, and asking about future directions beyond scaling.