AI 导读
过程分数不只是打分。它驱动一个修复循环。 当过程验证器将某次运行标记为有缺陷时,该诊断就变成一条修复提示。求解器在不看答案的情况下重试同一任务。 据 Apodex 报告的评估,在过程验证器标记为有缺陷的 434 条轨迹中,修复后的重跑平均得分高出 0.155。
正文
The process score does more than grade. It drives a repair loop.
When the process verifier flags a run as deficient, that diagnosis becomes a repair note. The solver retries the same task without seeing the answer.
Across 434 trajectories marked deficient by the process verifier, repaired reruns scored 0.155 higher on average, per Apodex's reported evaluation.
来源:@omarsar0 · x.com