AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

EarthVL: A Progressive Earth Vision-Language Understanding and Generation Framework

Junjue Wang, Yanfei Zhong, Zihang Chen, Zhuo Zheng, Ailong Ma, Liangpei Zhang

TPAMI 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 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 segmentation guidance (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 79.63 -> 81.06Improved
add segmentation guidance (ablation)EarthVLNet · EarthVLSet multiple-choice VQARMSE: 2.636 -> 2.340Improved
add segmentation guidance (ablation)EarthVLNet · EarthVLSet open-ended VQALAVE (%): 77.94 -> 80.44Improved
add segmentation guidance (ablation)EarthVLNet · EarthVLSet open-ended VQABLEU1: 0.5653 -> 0.5726Improved
add object-guided attention (OGA) instead of SE channel attention (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 80.72 -> 81.06Improved
add object-guided attention instead of spatial attention SA (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 79.83 -> 81.06Improved
add spatial attention SA on concatenated features (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 80.21 -> 79.83Hurt
add object guidance with OGA vs only features (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%) / OR: 79.97 -> 81.06; 2.582 -> 2.340Improved
use ND-Separated optimization instead of CE loss (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 79.91 -> 81.06Improved
use ND-Separated instead of ND-Shared (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 80.63 -> 81.06Improved
use DIW loss instead of CE loss (ablation)EarthVLNet · EarthVLSet multiple-choice VQAOA (%): 79.91 -> 79.51Hurt
scale up ConvNeXt vision encoder (ablation)Semantic-FPN + FlanT5-XL · EarthVLSet open-ended VQABLEU1: 0.505 -> 0.564Improved

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