fast method has bug when compute S1 conf/ST conf
fast method has bug when compute S1 conf/ST conf: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). this is my setting: when the code do fast analyze as following sentence: The bug is comming: In my opinion: It is np.arange(1, int((N + 1) / 2)) cause the bug, np.arange return…
The task
this is my setting: when the code do fast analyze as following sentence: The bug is comming: In my opinion: It is np.arange(1, int((N + 1) / 2)) cause the bug, np.arange return is [1,2,.....2467], however the lenth of f is 2467, the max index is 2466, So, I think maybe File…
Part of XiaomiMiMo/MiMo-V2.6-RL-oss.