3D-Aware Vision-Language Models Fine-Tuning with Geometric Distillation
This work addresses the problem of enhancing 3D understanding in VLMs for spatially grounded multimodal tasks, representing an incremental improvement by fine-tuning existing models.
The paper tackles the limitation of Vision-Language Models (VLMs) in understanding 3D spatial structures by proposing Geometric Distillation, a lightweight fine-tuning framework that injects geometric cues into pretrained VLMs, resulting in improved 3D spatial reasoning with significantly lower computational cost.
Vision-Language Models (VLMs) have shown remarkable performance on diverse visual and linguistic tasks, yet they remain fundamentally limited in their understanding of 3D spatial structures. We propose Geometric Distillation, a lightweight, annotation-free fine-tuning framework that injects human-inspired geometric cues into pretrained VLMs without modifying their architecture. By distilling (1) sparse correspondences, (2) relative depth relations, and (3) dense cost volumes from off-the-shelf 3D foundation models (e.g., MASt3R, VGGT), our method shapes representations to be geometry-aware while remaining compatible with natural image-text inputs. Through extensive evaluations on 3D vision-language reasoning and 3D perception benchmarks, our method consistently outperforms prior approaches, achieving improved 3D spatial reasoning with significantly lower computational cost. Our work demonstrates a scalable and efficient path to bridge 2D-trained VLMs with 3D understanding, opening up wider use in spatially grounded multimodal tasks.