Beyond the Trade-off: Self-Supervised Reinforcement Learning for Reasoning Models' Instruction Following

🤗 Upvotes: 25 | cs.AI

Authors:
Qingyu Ren, Qianyu He, Bowei Zhang, Jie Zeng, Jiaqing Liang, Yanghua Xiao, Weikang Zhou, Zeye Sun, Fei Yu

Title:
Beyond the Trade-off: Self-Supervised Reinforcement Learning for Reasoning Models' Instruction Following

Arxiv:
http://arxiv.org/abs/2508.02150v1

Abstract:
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.