通过随机探索提升推理模型表现，无需重训练即可显著提高解题准确率。
Boosting Inference with Guided Reasoning: Stochastic Exploration for Recursive Models
- 引入随机扰动机制，在推理时探索邻近推理路径。
- 在Sudoku-Extreme上将准确率从85.9%提升至98.0%。
- 提供三个无标签诊断指标，判断是否可改进及信任输出结果。
近期研究显示，小型神经网络在结构化推理任务中具有惊人潜力，关键在于用潜在动态系统建模推理轨迹。我们提出，此类架构的推理行为可视为对潜在推理轨迹的近似推断，确定性递归是单粒子、零噪声极限。通过引导式随机探索实现该观点：对推理动态施加随机扰动以生成邻近轨迹，模型原有的早停头在线重加权这些轨迹。该框架产生三个无标签诊断指标：局部稳定性、引导对齐度与云令牌熵，仅凭推理痕迹即可预测方法是否有效及哪些输出可信。在Sudoku-Extreme上，准确率从85.9%提升至98.0%，且无需重训练；在Maze-Hard中，诊断发现引导偏差，后续验证性能证实了该问题。该机制同时刻画了递归推理在轨迹层面是否有改进空间，以及模型内部引导能否修复问题。
原文摘要 · Abstract (English)
Recent work on recursive architectures has shown that tiny neural networks can be surprisingly powerful on structured reasoning tasks. The trick is to model reasoning trajectories with a latent dynamical system. We argue that the inference-time behaviour of these architectures is best understood as approximate inference over latent reasoning trajectories, with deterministic recursion as the one-particle, zero-noise limit. We make this view operational through guided stochastic exploration: stochastic perturbations of the reasoning dynamics propose neighbouring trajectories, and the model's existing early-stopping head reweights them online. The framework yields three label-free diagnostics: local stability, guide alignment, and cloud-token entropy. These predict, from inference traces alone, whether the procedure will help and which of its outputs to trust. On Sudoku-Extreme it lifts exact-solve accuracy from $85.9\%$ to $98.0\%$ without retraining; on Maze-Hard the diagnostics flag a misaligned guide, as validation performance later confirms. The same machinery thus characterises both when recursive reasoning has room to improve at the trajectory level and when the model's internal guide can recover it.
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。