Recursive Inference Machines for Neural Reasoning

Neural reasoners such as Tiny Recursive Models (TRMs) solve complex problems by combining neural backbones with specialized inference schemes. Such inference schemes have been a central component of stochastic reasoning systems, where inference rules are applied to a stochastic model to derive answers to complex queries. In this work, we bridge these two paradigms by introducing Recursive Inference Machines (RIMs), a neural reasoning framework that explicitly incorporates recursive inference mechanisms inspired by classical inference engines. We show that TRMs can be expressed as an instance of RIMs, allowing us to extend them through a reweighting component, yielding better performance on challenging reasoning benchmarks, including ARC-AGI-1, ARC-AGI-2, and Sudoku Extreme. Furthermore, we show that RIMs can be used to improve reasoning on other tasks, such as the classification of tabular data, outperforming TabPFNs.

• Publication year

  2026

• Venue

  Unknown venue

• Publication date

  2026-03-05

• Fields of study

  Computer Science

• Identifiers
  arXiv 2603.05234
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Hierarchical Reasoning Model

  2025influential reference
• Universal Reasoning Model

  2025cited by this paper
• Tiny Recursive Models on ARC-AGI-1: Inductive Biases, Identity Conditioning, and Test-Time Compute

  2025cited by this paper
• Accurate predictions on small data with a tabular foundation model

  2025cited by this paper
• ARC-AGI-2: A New Challenge for Frontier AI Reasoning Systems

  2025cited by this paper
• Trust, But Verify: A Self-Verification Approach to Reinforcement Learning with Verifiable Rewards

  2025cited by this paper
• Autoregressive Large Language Models are Computationally Universal

  2024cited by this paper
• Beyond A*: Better Planning with Transformers via Search Dynamics Bootstrapping

  2024cited by this paper
• A Unified Framework for Human-Allied Learning of Probabilistic Circuits

  2024cited by this paper
• xLSTM: Extended Long Short-Term Memory

  2024cited by this paper
• Neuro-symbolic artificial intelligence: a survey

  2024cited by this paper
• RWKV: Reinventing RNNs for the Transformer Era

  2023cited by this paper
• Mamba: Linear-Time Sequence Modeling with Selective State Spaces

  2023cited by this paper
• Tree of Thoughts: Deliberate Problem Solving with Large Language Models

  2023cited by this paper
• Chain of Thought Prompting Elicits Reasoning in Large Language Models

  2022cited by this paper
• Self-Consistency Improves Chain of Thought Reasoning in Language Models

  2022cited by this paper
• SATformer: Transformer-Based UNSAT Core Learning

  2022cited by this paper
• Lattice Recurrent Unit: Improving Convergence and Statistical Efficiency for Sequence Modeling

  2017cited by this paper
• On the Properties of Neural Machine Translation: Encoder–Decoder Approaches

  2014cited by this paper
• A measure of intelligence

  2012cited by this paper
• Learning message-passing inference machines for structured prediction

  2011cited by this paper
• Probabilistic graphical models : principles and techniques

  2009cited by this paper
• Probabilistic reasoning in intelligent systems: Networks of plausible inference

  1991cited by this paper
• Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images

  1984cited by this paper
• Reverend Bayes on Inference Engines: A Distributed Hierarchical Approach

  1982cited by this paper

• No citing papers are available for this paper.