Fine-Grained Preference Optimization Improves Spatial Reasoning in VLMs
Abstract
SpatialReasoner-R1, a vision-language reasoning model, uses Multi-Model Monte Carlo Tree Search and fine-grained Direct Preference Optimization to improve spatial reasoning, setting a new state-of-the-art on SPATIALRGPT-Bench.
Current Vision-Language Models (VLMs) struggle with fine-grained spatial reasoning, particularly when multi-step logic and precise spatial alignment are required. In this work, we introduce SpatialReasoner-R1, a vision-language reasoning model designed to address these limitations. To construct high-quality supervision for spatial reasoning, we design a Multi-Model Monte Carlo Tree Search (M3CTS) method that generates diverse, logically consistent Long Chain-of-Thought (LongCoT) reasoning trajectories. In addition, we propose fine-grained Direct Preference Optimization (fDPO), which introduces segment-specific preference granularity for descriptive grounding and logical reasoning, guided by a spatial reward mechanism that evaluates candidate responses based on visual consistency, spatial grounding, and logical coherence. Experimental results demonstrate that fDPO achieves an average improvement of 4.1% over standard DPO across spatial quality tasks, and a 9.0% gain in spatial quantity tasks. SpatialReasoner-R1, trained with fDPO, sets a new SoTA on SPATIALRGPT-Bench, outperforming the strongest baseline by 9.8% in average accuracy, while maintaining competitive performance on general vision-language tasks.
Community
We propose a novel fine-grained preference optimization approach that significantly improves spatial reasoning capabilities in Vision-Language Models (VLMs). Our method leverages carefully designed preference data and training strategies to enhance spatial understanding without compromising general visual capabilities.
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper