Research
GeoBridge-VLA: Geometry-Aware Residual Adaptation for Vision-Language-Action Models
arXiv:2610.05026v1 Announce Type: new Abstract: Vision-language-action (VLA) models encode semantic information from vision-language pretraining, but manipulation also requires precise spatial reasoni
arXiv:2610.05026v1 Announce Type: new Abstract: Vision-language-action (VLA) models encode semantic information from vision-language pretraining, but manipulation also requires precise spatial reasoning. We present GeoBridge-VLA, a two-stage method for learning geometric features from a pretrained VLA's frozen visual encoder and using them for action prediction. Stage I trains a feature bridge and geometry decoder with depth supervision. Stage II freezes these modules and trains a gated residual interface together with the action-side projections and action expert. The residual augments the existing visual tokens without adding a second image encoder or increasing the token count. Deployment requires RGB, robot state, and language, but no depth observations. Under matched evaluation conditions, GeoBridge-VLA achieves 70.9% success on LIBERO, compared with 60.0% for SmolVLA. Disabling the residual in the same trained checkpoint reduces success from 70.90% to 69.85%, with mixed effects across suites. On a physical ROBOTIS OMY robot, GeoBridge-VLA succeeds in 148 of 200 trials (74.0%) across four tasks, compared with 108 of 200 (54.0%) for SmolVLA.
Related
- StemVLA:An Open-Source Vision-Language-Action Model with Future 3D Spatial Geometry Knowledge and 4D Historical Representation
- 3DVLA: Enhancing Vision-Language-Action Models via 3D Spatial and Instance Understanding
- Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning
- Prompt-Anchored Residual Adaptation for Biomedical Vision-Language Models
Source: arXiv cs.CV | 2026-10-06