HF RL Explorer

The model YAML/dict persistence API needs to be reliable for workflows that repeatedly serialize model…

The model YAML/dict persistence API needs to be reliable for workflows that repeatedly serialize model…: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). At present, serializing a collection can mutate live Parameter objects and produce different link identifiers on each…

The task

At present, serializing a collection can mutate live Parameter objects and produce different link identifiers on each call, while malformed or inconsistent link metadata can be silently accepted during loading. This makes generated configuration hard to diff and can turn a typo…

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