Qwen2.5-0.5B

1 article tagged with this entity.

Reward Granularity in RLVR: Comparing Process and Outcome Reward Structures for Mathematical Reasoning in Small Language Models