AI RESEARCH

SpaceMind++: Toward Allocentric Cognitive Maps for Spatially Grounded Video MLLMs

arXiv CS.CV

ArXi:2605.09449v1 Announce Type: new Recent multimodal large language models (MLLMs) have made remarkable progress in visual understanding and language-based reasoning, yet they lack a persistent world-centered representation for spatially consistent reasoning in 3D environments. Inspired by the mammalian dual-stream system, where semantic and spatial cues are processed separately and integrated into an allocentric cognitive map, we propose SpaceMind++, a video MLLM architecture that explicitly builds a voxelized cognitive map from RGB videos.