INTCORT: Training-Free Spatial Reasoning Enhancement for Vision-Language Models via Input Transformations and Confidence Routing

2026-09-21

AI Summary

本文针对视觉-语言模型空间推理能力差的问题，提出INTCORT框架，通过输入变换和关系令牌置信度路由增强空间推理能力，无需训练且不影响模型通用性能。

Abstract

Vision-Language Models (VLMs) have demonstrated remarkable capabilities in multimodal tasks, yet they still exhibit poor ability in spatial reasoning. Existing training-dependent and training-free enhancement methods suffer from high computational costs with catastrophic forgetting and internal mechanism interference that compromises general capabilities, respectively. In this work, we first verify two key hypotheses: appropriate geometric image transformation and query-reversal transformation can recover incorrect spatial predictions, and correct predictions exhibit higher relation-token confidence than incorrect ones. Based on these findings, we propose INTCORT, a training-free spatial reasoning enhancement framework that constructs multiple inference views through input transformations and aggregates their predictions via relation-token confidence routing, without modifying the VLM's internal mechanisms. Experimental results on several commonly-used benchmarks demonstrate that INTCORT substantially improves spatial reasoning accuracy across diverse VLMs, achieving an average improvement of 10.01% over all models and benchmarks. Compared with prior works, INTCORT achieves superior performance with improvements of up to 25.01%.

Problem

• Vision-Language Models
• spatial reasoning
• computational costs
• catastrophic forgetting
• internal mechanism interference

Innovation

• Training-Free
• Spatial Reasoning
• Input Transformations
• Confidence Routing