DRQA: Dynamic Reasoning Quota Allocation for Controlling Overthinking in Reasoning Large Language Models

Open in new window