让模型在隐空间中稳定迭代推理，提升数学逻辑题解题能力
SpiralThinker: Latent Reasoning through an Iterative Process with Text-Latent Interleaving
- 隐空间与文本推理交替进行，通过逐步对齐稳定更新
- 在数学逻辑任务上超越现有隐式推理方法
- 适合需要深度推理的复杂问题求解场景
近期大模型推理进展依赖强化学习与测试时扩展，研究重点转向隐空间而非纯文本推理。然而，现有隐式推理方法缺乏保证隐空间推理稳定性机制，也缺少显式与隐式推理系统性交织方式。本文提出SpiralThinker，一种通过迭代更新隐表示并交错进行隐式与显式推理的稳定框架。核心在于引入渐进对齐目标，显式调控多轮迭代中的隐表示，并通过结构化标注实现文本-隐空间交互，从而稳定隐空间更新并保持与文本推理的一致性。在数学、逻辑及常识推理任务上，SpiralThinker优于所有现有隐式推理基线。进一步分析表明：迭代与对齐均至关重要；最优隐向量数与迭代次数随数据集变化；恰当对齐是高效迭代隐式推理的关键。总体而言，SpiralThinker连接了迭代计算与隐式推理，证明对齐的迭代更新可可靠引导隐空间推理。
原文摘要 · Abstract (English)
Recent advances in large reasoning models have been driven by reinforcement learning and test-time scaling, accompanied by growing interest in latent rather than purely textual reasoning. However, existing latent reasoning methods lack mechanisms to ensure stable reasoning dynamics in latent space and a systematic way to interleave implicit and explicit reasoning. We introduce SpiralThinker, a stabilized iterative latent reasoning framework that performs iterative updates over latent representations while interleaving latent and textual reasoning steps. At its core, it combines a progressive alignment objective that explicitly regulates latent representations across iterations with structured annotations for text-latent interleaving, thereby stabilizing latent updates and maintaining coherence with textual reasoning. Across mathematical, logical, and commonsense reasoning tasks, SpiralThinker achieves state-of-the-art performance among latent reasoning baselines. Further analysis shows that both iteration and alignment are essential, that the optimal numbers of latent tokens and iterations vary by dataset, and that proper alignment is crucial for effective iterative latent reasoning. Overall, SpiralThinker bridges iterative computation and latent reasoning, demonstrating that aligned iterative updates can reliably steer reasoning in the latent space.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。