HF RL Explorer

qingyuanwu/reconcile-workflow-lab

reconcile-workflow-lab: RL dataset on Hugging Face. A self-contained OpenEnv environment with 48 original synthetic tasks: six per domain, covering all three difficulty levels. Each frozen identifier selects the same case across fresh Arena containers and repeated rollouts. Selection happens…

qingyuanwu/reconcile-workflow-lab on the Hugging Face Hub