rSIM: Incentivizing Reasoning Capabilities of LLMs via Reinforced Strategy Injection
This addresses the challenge of improving reasoning in LLMs for AI applications, though it appears incremental as it builds on existing chain-of-thought and reinforcement learning methods.
The paper tackles the problem of enhancing reasoning capabilities in large language models (LLMs) by introducing a reinforced strategy injection mechanism (rSIM) that uses a small planner trained with multi-agent RL to guide reasoning strategies, resulting in a 0.5B model outperforming a 14B model.
Large language models (LLMs) are post-trained through reinforcement learning (RL) to evolve into Reasoning Language Models (RLMs), where the hallmark of this advanced reasoning is ``aha'' moments when they start to perform strategies, such as self-reflection and deep thinking, within chain of thoughts (CoTs). Motivated by this, this paper proposes a novel reinforced strategy injection mechanism (rSIM), that enables any LLM to become an RLM by employing a small planner to guide the LLM's CoT through the adaptive injection of reasoning strategies. To achieve this, the planner (leader agent) is jointly trained with an LLM (follower agent) using multi-agent RL (MARL), based on a leader-follower framework and straightforward rule-based rewards. Experimental results show that rSIM enables Qwen2.5-0.5B to become an RLM and significantly outperform Qwen2.5-14B. Moreover, the planner is generalizable: it only needs to be trained once and can be applied as a plug-in to substantially improve the reasoning capabilities of existing LLMs. In addition, the planner supports continual learning across various tasks, allowing its planning abilities to gradually improve and generalize to a wider range of problems.