RISER: 通过潜意识推理技能协调实现自适应激活引导
RISER: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering
AI总结：
RISER通过构建可重用的推理向量库和轻量级路由器，实现自适应激活引导，提升LLM推理的准确率和效率。
摘要：
近期基于大语言模型（LLMs）的领域特定推理研究往往依赖于训练密集型方法，需要参数更新。虽然激活引导已经作为一种参数高效的替代方法出现，但现有方法应用静态的手动干预，无法适应复杂推理的动态特性。为了解决这一限制，我们提出了RISER（基于路由的可引导推理增强干预），一种即插即用的干预框架，能够自适应地在激活空间中引导LLM推理。RISER构建了一个可重用的推理向量库，并采用轻量级的路由器动态组合每个输入的向量。路由器通过任务层面的奖励进行优化，以涌现和组合的方式激活潜在的认知原语。在七个多样化的基准测试中，RISER在零样本准确率上平均提高了3.4-6.5%，超过CoT式推理，并在2-3倍更高的token效率和鲁棒准确率提升方面超越。进一步分析显示，RISER能够自动将多个向量组合成可解释、精确的控制策略，指向更加可控和高效的LLM推理。
英文摘要：
Recent work on domain-specific reasoning with large language models (LLMs) often relies on training-intensive approaches that require parameter updates. While activation steering has emerged as a parameter efficient alternative, existing methods apply static, manual interventions that fail to adapt to the dynamic nature of complex reasoning. To address this limitation, we propose RISER (Router-based Intervention for Steerable Enhancement of Reasoning), a plug-and-play intervention framework that adaptively steers LLM reasoning in activation space. RISER constructs a library of reusable reasoning vectors and employs a lightweight Router to dynamically compose them for each input. The Router is optimized via reinforcement learning under task-level rewards, activating latent cognitive primitives in an emergent and compositional manner. Across seven diverse benchmarks, RISER yields 3.4-6.5% average zero-shot accuracy improvements over the base model while surpassing CoT-style reasoning with 2-3x higher token efficiency and robust accuracy gains. Further analysis shows that RISER autonomously combines multiple vectors into interpretable, precise control strategies, pointing toward more controllable and efficient LLM reasoning.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
