我们来看看第三个评价任务，以免您认为这是因为随机数生成器的性能有所不同。

Lest you think that this is due to the performance of RNGs, let's take a look at this third evaluative task.

youdao

利用随机生成的任务交互图进行测试，验证了模型的正确性以及算法的有效性。

At last, we use randomly generated task interaction graph to test the validity of the model and algorithm.

youdao