AI Paper Atlas

Research map / Vision-language and multimodal / Cross-modal learning

MaRS: A Multi-modality Very-high-resolution Remote Sensing Foundation Model with Cross-Granularity Meta-Modality Learning

Ruoyu Yang, Yinhe Liu, Heng Yan, Yiheng Zhou, Yihan Fu, Han Luo, Yanfei Zhong

AAAI 2026 · Vision-language and multimodal › Cross-modal learning · approach: Multi-modality image fusion · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add CGCL to the MIM + contrastive baseline (ablation)MaRS baseline · EarthMiss modality-missing mappingmIoU: 42.0 -> 49.1Improved
add CGCL to the MIM + contrastive baseline (ablation)MaRS baseline · WHU-CD change detectionIoU: 67.1 -> 76.0Improved
add MMA to the MIM + contrastive baseline (ablation)MaRS baseline · WHU-CD change detectionIoU: 67.1 -> 86.2Improved
add MMA to the MIM + contrastive baseline (ablation)MaRS baseline · EarthMiss modality-missing mappingmIoU: 42.0 -> 48.4Improved
add MMA to the baseline (registration) (ablation)MaRS baseline · GUSO registrationRMSE: 2.835 -> 2.854Hurt
add CGCL to the baseline (registration) (ablation)MaRS baseline · GUSO registrationRMSE: 2.835 -> 2.795Improved
combine CGCL and MMA (full MaRS) vs either alone (ablation)MaRS · SARDet-100KmAP: 53.5 -> 55.4Improved
combine CGCL and MMA vs CGCL alone (ablation)MaRS · WHU-CD change detectionIoU: 76.0 -> 87.0Improved
use hierarchical SwinV2 backbone instead of ViT-LVFM baselines · Multi-modal VHR tasksvarious: Building height Delta 23.26 -> 23.38Improved
pretrain on medium-resolution data (RSFMs) rather than VHR dataRSFM baselines · Multi-modal VHR tasksvarious: e.g. EarthMiss mIoU SatMAE 35.0 vs MaRS 49.9Hurt
pretrain MaRS on MaRS-16M vs generic SwinV2SwinV2-B · GUSO translationPSNR: 16.71 -> 20.69Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy