Boosting Inference with Guided Reasoning: Stochastic Exploration for Recursive Models

Andrew Corbett, Archit Sood, Anna Tzatzopoulou, Sai-Aakash Ramesh, Tim James Dodwell

Keywords: Tiny Recursive Models, Approximate Inference, Latent Reasoning Models, Sequential Monte Carlo

TL;DR: Tiny recursive models know more than they let on: embedding them as a deterministic trajectory in an approximate inference framework lifts Sudoku-Extreme accuracy from 85.9% to 98.0% with no retraining, plus diagnostics flag when the boost kicks in.

Abstract: 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.

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 74