AI RESEARCH
GAP-MLLM: Geometry-Aligned Pre-training for Activating 3D Spatial Perception in Multimodal Large Language Models
arXiv CS.CV
•
ArXi:2603.16461v1 Announce Type: new Multimodal Large Language Models (MLLMs) nstrate exceptional semantic reasoning but struggle with 3D spatial perception when restricted to pure RGB inputs. Despite leveraging implicit geometric priors from 3D reconstruction models, image-based methods still exhibit a notable performance gap compared to methods using explicit 3D data. We argue that this gap does not arise from insufficient geometric priors, but from a misalignment in the