The field of spatial intelligence in multimodal models is rapidly advancing, with a focus on improving the ability of models to understand and reason about 3D spatial relationships. Recent work has highlighted the importance of decoupling 3D reasoning from numerical regression, and has introduced novel architectures and benchmarks to support this goal. Notable papers in this area include Beyond Flatlands, which introduces a new architecture for spatial intelligence, and GGBench, which provides a comprehensive benchmark for evaluating geometric generative reasoning. Other notable papers include Video Spatial Reasoning with Object-Centric 3D Rollout, GeoX-Bench, and Cognitive Maps in Language Models. These papers demonstrate significant advancements in spatial intelligence, including improved performance on benchmarks and the development of new methods for spatial reasoning.
Spatial Intelligence in Multimodal Models
Sources
Beyond Flatlands: Unlocking Spatial Intelligence by Decoupling 3D Reasoning from Numerical Regression
GGBench: A Geometric Generative Reasoning Benchmark for Unified Multimodal Models
Video Spatial Reasoning with Object-Centric 3D Rollout
GeoX-Bench: Benchmarking Cross-View Geo-Localization and Pose Estimation Capabilities of Large Multimodal Models
Cognitive Maps in Language Models: A Mechanistic Analysis of Spatial Planning
Scaling Spatial Intelligence with Multimodal Foundation Models
Imagine in Space: Exploring the Frontier of Spatial Intelligence and Reasoning Efficiency in Vision Language Models
Spatial Reasoning in Multimodal Large Language Models: A Survey of Tasks, Benchmarks and Methods
Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoning
StreetView-Waste: A Multi-Task Dataset for Urban Waste Management
Solving Spatial Supersensing Without Spatial Supersensing