Model Releases
From Sight to Foresight: Predictive Spatial Reasoning in Vision-Language Models
arXiv:2610.04139v1 Announce Type: new Abstract: Predicting future spatial states supports collision avoidance and timely decision-making in dynamic environments. However, existing vision-language mode
arXiv:2610.04139v1 Announce Type: new Abstract: Predicting future spatial states supports collision avoidance and timely decision-making in dynamic environments. However, existing vision-language models (VLMs) and benchmarks for spatial reasoning primarily focus on observed scenes, leaving predictive spatial reasoning beyond the observed interval underexplored. To this end, we introduce SpatialMind, a metric-scale VLM for spatial reasoning and future prediction. Its metric depth adapter anchors spatial reasoning to real-world scale, while its progressive state chain establishes current spatial states and observed dynamics as the foundation for future prediction. Given a video prefix, SpatialMind predicts distances, motion directions, and spatial relations in both observed and unseen future frames. For training and evaluation, we build a scalable data engine that grounds entity descriptions in metric geometry to generate question-answer pairs and state supervision. Using this engine, we construct the SpatialMind-30K dataset and the SpatialMind-2K benchmark, both covering driving and everyday egocentric scenes. The benchmark spans eight tasks across three levels: current-state understanding, observed-dynamics understanding, and future prediction. Experiments show that SpatialMind substantially outperforms both general and spatially specialized models on our benchmark while achieving competitive zero-shot performance on VSI-Bench, OSI-Bench, and VLM4D.
Related
- CL4D: Contrastive Language-4D Pretraining for Vision-Language Reasoning in Dynamic Scenes
- EgoPathBench: Evaluating Zero-Shot Egocentric Waypoint Decision-Making in Vision-Language Models
- SpaMEM: Benchmarking Dynamic Spatial Reasoning via Perception-Memory Integration in Embodied Environments
- LaST-R1: Reinforcing Action via Adaptive Physical Latent Reasoning for VLA Models
- CAR-VLA: Complexity-Aware and Risk-Adaptive Reasoning for Autonomous Driving
Source: arXiv cs.CV | 2026-10-06