Beyond the Trade-off: Self-Supervised Reinforcement Learning for Reasoning Models' Instruction Following
This addresses a methodological bottleneck in enhancing instruction following for reasoning models, offering a scalable and cost-effective solution.
The paper tackles the trade-off between reasoning capabilities and instruction following in reasoning models by introducing a self-supervised RL framework that uses internal signals to improve instruction following without external supervision, achieving significant improvements while maintaining reasoning performance.
Reasoning models excel in complex problem solving but exhibit a concerning trade off between reasoning capabilities and instruction following abilities. Existing approaches for improving instruction following rely on stronger external models, creating methodological bottlenecks and practical limitations including increased costs and accessibility constraints. We propose a self-supervised RL framework that leverages reasoning models' own internal signals to improve instruction following capabilities without external supervision. Extensive experiments demonstrate that our framework significantly improves instruction following capabilities while maintaining reasoning performance, offering a scalable and cost-effective approach to enhance instruction following in reasoning models. The data and code are publicly available at https://github.com/Rainier-rq/verl-if.