HF RL Explorer

Multi-document YAML streams are often assembled from Kubernetes-style resources, and their document order is…

Multi-document YAML streams are often assembled from Kubernetes-style resources, and their document order is…: a task in MiMo-V2.6-RL-oss: Agentic RL Environments (MiMo RL release). Today dyff between compares documents only by position, so a harmless reorder produces misleading cross-resource…

The task

Today dyff between compares documents only by position, so a harmless reorder produces misleading cross-resource diffs, while streams with different document counts cannot be compared at all. Add an opt-in identity mode to dyff between through a repeatable --match-documents-by…

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