HF RL Explorer

I’m using ModelParamTraversal to walk and update parameter trees, but importing it from flax.optim feels…

I’m using ModelParamTraversal to walk and update parameter trees, but importing it from flax.optim feels…: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). Problem Statement I’m using ModelParamTraversal to walk and update parameter trees, but importing it from flax.optim…

The task

Problem Statement I’m using ModelParamTraversal to walk and update parameter trees, but importing it from flax.optim feels like the wrong place. Please make it available from flax.traverse_util as well, and move the docs there so it’s easier to find. Expected outcomes…

Part of XiaomiMiMo/MiMo-V2.6-RL-oss.