Several correlated but not strictly identical changes happened over the same few months: scaling inference compute, heavier use of RL in post-training, and models producing reasoning tokens.
大多数人可能将AI进步归因于单一因素(如模型规模或数据量),但作者指出推理能力的提升是多种因素共同作用的结果,包括推理计算扩展、强化学习更广泛应用以及模型产生推理标记等。这挑战了人们对AI进步驱动因素的认知。