Probabilistic Tiny Recursive Model
Amin Sghaier ⋅ Ali Parviz ⋅ Alexia Jolicoeur-Martineau
Abstract
Tiny Recursive Models (TRM) solve complex reasoning tasks with a fraction of the parameters of modern large language models (LLMs) by iteratively refining a latent state and final answer. While powerful, their deterministic recursion can lead to convergence at suboptimal solutions, without escape mechanism. A common workaround relies on task-specific input perturbations at test time combined with answer aggregation via voting. We introduce Probabilistic TRM (PTRM), a task-agnostic framework for test-time compute scaling that addresses this limitation through stochastic exploration. PTRM injects Gaussian noise at each deep recursion step, enabling parallel trajectories to explore diverse solution basins, and selects among them with a criterion such as the model’s existing Q head (used for early stopping in the original TRM). Without requiring retraining or task-specific augmentations, PTRM enables substantial accuracy gains across benchmarks, including Sudoku-Extreme ($87.28\%$ to $98.75\%$), N-Queens ($48.1\%$ to $75.5\%$ on 16×16 boards), and various puzzles from Pencil Puzzle Bench ($62.6\%$ to $91.2\%$). On the latter, PTRM achieves nearly double the accuracy of an ensemble of frontier LLMs ($91.2\%$ vs. $55.1\%$) at a small fraction of the cost, using only 7M parameters.
Successful Page Load