From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Oi, Masanari, Maeda, Koki, Koike, Ryuto, Oba, Daisuke, Inoue, Nakamasa, Okazaki, Naoaki
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917262730985472
author Oi, Masanari
Maeda, Koki
Koike, Ryuto
Oba, Daisuke
Inoue, Nakamasa
Okazaki, Naoaki
author_facet Oi, Masanari
Maeda, Koki
Koike, Ryuto
Oba, Daisuke
Inoue, Nakamasa
Okazaki, Naoaki
contents While multimodal large language models (MLLMs) have made substantial progress in single-image spatial reasoning, multi-image spatial reasoning, which requires integration of information from multiple viewpoints, remains challenging. Cognitive studies suggest that humans address such tasks through two mechanisms: cross-view correspondence, which identifies regions across different views that correspond to the same physical locations, and stepwise viewpoint transformation, which composes relative viewpoint changes sequentially. However, existing studies incorporate these mechanisms only partially and often implicitly, without explicit supervision for both. We propose Human-Aware Training for Cross-view correspondence and viewpoint cHange (HATCH), a training framework with two complementary objectives: (1) Patch-Level Spatial Alignment, which encourages patch representations to align across views for spatially corresponding regions, and (2) Action-then-Answer Reasoning, which requires the model to generate explicit viewpoint transition actions before predicting the final answer. Experiments on three benchmarks demonstrate that HATCH consistently outperforms baselines of comparable size by a clear margin and achieves competitive results against much larger models, while preserving single-image reasoning capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2602_08735
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Models
Oi, Masanari
Maeda, Koki
Koike, Ryuto
Oba, Daisuke
Inoue, Nakamasa
Okazaki, Naoaki
Computer Vision and Pattern Recognition
While multimodal large language models (MLLMs) have made substantial progress in single-image spatial reasoning, multi-image spatial reasoning, which requires integration of information from multiple viewpoints, remains challenging. Cognitive studies suggest that humans address such tasks through two mechanisms: cross-view correspondence, which identifies regions across different views that correspond to the same physical locations, and stepwise viewpoint transformation, which composes relative viewpoint changes sequentially. However, existing studies incorporate these mechanisms only partially and often implicitly, without explicit supervision for both. We propose Human-Aware Training for Cross-view correspondence and viewpoint cHange (HATCH), a training framework with two complementary objectives: (1) Patch-Level Spatial Alignment, which encourages patch representations to align across views for spatially corresponding regions, and (2) Action-then-Answer Reasoning, which requires the model to generate explicit viewpoint transition actions before predicting the final answer. Experiments on three benchmarks demonstrate that HATCH consistently outperforms baselines of comparable size by a clear margin and achieves competitive results against much larger models, while preserving single-image reasoning capabilities.
title From Correspondence to Actions: Human-Like Multi-Image Spatial Reasoning in Multi-modal Large Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.08735