Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces
This addresses the bottleneck of spatial reasoning in MLLMs for applications like robotics and navigation, though it is incremental as it builds on existing models and benchmarks.
The paper tackled the problem of whether Multimodal Large Language Models (MLLMs) can think in space from videos, finding that they exhibit competitive but subhuman visual-spatial intelligence on a new benchmark of over 5,000 question-answer pairs, with explicit generation of cognitive maps improving spatial distance ability.
Humans possess the visual-spatial intelligence to remember spaces from sequential visual observations. However, can Multimodal Large Language Models (MLLMs) trained on million-scale video datasets also ``think in space'' from videos? We present a novel video-based visual-spatial intelligence benchmark (VSI-Bench) of over 5,000 question-answer pairs, and find that MLLMs exhibit competitive - though subhuman - visual-spatial intelligence. We probe models to express how they think in space both linguistically and visually and find that while spatial reasoning capabilities remain the primary bottleneck for MLLMs to reach higher benchmark performance, local world models and spatial awareness do emerge within these models. Notably, prevailing linguistic reasoning techniques (e.g., chain-of-thought, self-consistency, tree-of-thoughts) fail to improve performance, whereas explicitly generating cognitive maps during question-answering enhances MLLMs' spatial distance ability.