MindCube: Spatial Mental Modeling from Limited Views

TL;DR

This paper introduces the MindCube benchmark to evaluate vision-language models' ability to form spatial mental models from limited views, and proposes a 'map-then-reason' approach that significantly improves performance.

Contribution

The paper presents a new benchmark and a novel training approach that enhances VLMs' capacity to build and reason over internal spatial representations.

Findings

VLMs perform near-random on the MindCube benchmark.

The 'map-then-reason' approach improves accuracy from 37.8% to 57.8%.

Reinforcement learning further boosts performance to 61.3%.

Abstract

Can Vision-Language Models (VLMs) imagine the full scene from just a few views, like humans do? Humans form spatial mental models naturally, internal representations of unseen space, to reason about layout, perspective, and motion. Our MindCube benchmark with 21,154 questions across 3,268 images exposes this critical gap, where existing VLMs exhibit near-random performance. Using MindCube, we systematically evaluate how well VLMs build robust spatial mental models through representing positions (cognitive mapping), orientations (perspective-taking), and dynamics (mental simulation for "what-if" movements). We then explore three approaches to help approximate spatial mental models in VLMs, focusing on incorporating unseen intermediate views, natural language reasoning chains, and cognitive maps. The significant improvement comes from a synergistic approach, "map-then-reason", that jointly trains the model to first generate a cognitive map and then reason upon it. By training models to reason over these internal maps, we boosted accuracy from 37.8% to 57.8% (+20.0%). Adding reinforcement learning pushed performance even further to 61.3% (+23.5%). Our key insight is that such scaffolding of spatial mental models, actively constructing and utilizing internal structured spatial representations with flexible reasoning processes, significantly improves understanding of unobservable space.