BROWSE
Detail
Conference Paper
MultihopSpatial: Multi-hop Compositional Spatial Reasoning Benchmark for Vision-Language Model
Cited - time in 
Share 
Share
- Authors
- Youngwan Lee, Soojin Jang, Yoorhim Cho, Seunghwan Lee, Yong-Ju Lee, Sung Ju Hwang
- Issue Date
- 2026-09
- Citation
- European Conference on Computer Vision (ECCV) 2026, pp.1-18
- Publisher
- Springer
- Language
- English
- Type
- Conference Paper
- Abstract
- Spatial reasoning is foundational for Vision-Language Mod-els (VLMs), particularly when deployed as Vision-Language-Action (VLA)agents in physical environments. However, existing benchmarks predomi-nantly focus on elementary, single-hop relations, neglecting the multi-hopcompositional reasoning and precise visual grounding essential for real-world scenarios. To address this, we introduce MultihopSpatial, o!er-ing three key contributions: (1) A comprehensive benchmark designedfor multi-hop and compositional spatial reasoning, featuring 1- to 3-hopcomplex queries across diverse spatial perspectives. (2) Acc@50IoU,a complementary metric that simultaneously evaluates reasoning andvisual grounding by requiring both answer selection and precise bound-ing box prediction—capabilities vital for robust VLA deployment. (3)MultihopSpatial-Train, a dedicated large-scale training corpus to fos-ter spatial intelligence. Extensive evaluation of 37 state-of-the-art VLMsyields eight key insights, revealing that compositional spatial reasoningremains a formidable challenge. Finally, we demonstrate that reinforce-ment learning post-training on our corpus enhances both intrinsic VLMspatial reasoning and downstream embodied manipulation performance.
- Keyword
- Vision Foundation Models: Interpretability and Reasoning
- KSP Keywords
- Answer selection, Extensive evaluation, Language Models, Manipulation Performance, Multi-Hop, Real-world, Spatial intelligence, Spatial reasoning, large-scale, single-hop, state-of-The-Art