Spatial Reasoners for Continuous Variables in Any Domain

TL;DR

Spatial Reasoners is a flexible software framework that enables reasoning over continuous variables using generative denoising models, simplifying research in this emerging area.

Contribution

It provides an easy-to-use interface for integrating various generative models, inference strategies, and data domains for continuous variable reasoning.

Findings

Facilitates research in generative reasoning over continuous variables

Supports multiple denoising formulations and inference strategies

Open-source availability encourages community adoption

Abstract

We present Spatial Reasoners, a software framework to perform spatial reasoning over continuous variables with generative denoising models. Denoising generative models have become the de-facto standard for image generation, due to their effectiveness in sampling from complex, high-dimensional distributions. Recently, they have started being explored in the context of reasoning over multiple continuous variables. Providing infrastructure for generative reasoning with such models requires a high effort, due to a wide range of different denoising formulations, samplers, and inference strategies. Our presented framework aims to facilitate research in this area, providing easy-to-use interfaces to control variable mapping from arbitrary data domains, generative model paradigms, and inference strategies. Spatial Reasoners are openly available at https://spatialreasoners.github.io/