HF RL Explorer

Source freshness state selection is used to defer work until an upstream source has actually advanced, but…

Source freshness state selection is used to defer work until an upstream source has actually advanced, but…: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). Make the public source status:fresher selector compare the current and previous freshness artifacts deterministically…

The task

Make the public source_status:fresher selector compare the current and previous freshness artifacts deterministically. A source is fresher when it has no matching entry in the previous artifact, or when its current max_loaded_at is strictly later than the previous value. Equal…

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