Control-R: Towards controllable test-time scaling

The paper introduces Reasoning Control Fields (RCF), a novel test-time method that injects structured control signals to guide large reasoning models in complex tasks, addressing issues of under- and overthinking in chain-of-thought reasoning. It also presents the Control-R-4K dataset with annotated reasoning processes and proposes a Conditional Distillation Finetuning (CDF) approach, achieving state-of-the-art results and enabling controllable, scalable reasoning at the 32B model scale.