AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

GeoReason: Bridging Logical Reasoning and Spatial Fidelity in Remote Sensing Segmentation

Cheng Fan, Han Wu, Tingzhang Luo, Minjing Dong, Yehui Tang, Jianyuan Guo

NeurIPS 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Spatial reasoning in VLMs · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes GeoReason, an MLLM-based approach for instruction-driven reasoning segmentation in remote sensing imagery, aimed at fixing spatial fidelity degradation and target misidentification in the Embedding-as-Mask paradigm.

Problem

Reasoning segmentation of remote sensing imagery with MLLMs under cluttered geographic contexts and extreme scale variation, which causes spatial fidelity degradation and target misidentification.

Approach

Extends the Embedding-as-Mask paradigm of MLLM-based segmentation to remote sensing reasoning segmentation; the abstract does not specify the mechanism beyond targeting spatial fidelity and logical reasoning.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add Anticipatory Prior (prompt-side [SEG] token) (ablation)GeoReason · EarthReason testcIoU: 70.45 -> 78.87Improved
add Latent PTA loss (ablation)GeoReason · EarthReason testcIoU: 70.45 -> 73.01Improved
add Location-aware CoT (BBox) alone, without AP (ablation)GeoReason · EarthReason testgIoU / cIoU: gIoU 75.32 -> 74.54; cIoU 70.45 -> 70.59No clear effect
add BBox alone (ablation)GeoReason · RRSIS-D testgIoU: 66.42 -> 65.29Hurt
add PTA loss (ablation)GeoReason · RRSIS-D testgIoU: 66.42 -> 66.99Improved
add BBox and PTA on top of AP (full model) (ablation)GeoReason · EarthReason testcIoU: 78.87 -> 80.60Improved
place extra [SEG] in the answer stream instead of prompt (ablation)GeoReason · EarthReasonAvg of gIoU and cIoU: 78.61 -> 74.14Hurt
remove CoT reasoning (Zero-Reasoning) from baseline (ablation)Baseline · EarthReasonAvg of gIoU and cIoU: 72.79 -> 74.63Improved
remove reasoning from GeoReason (ablation)GeoReason · EarthReasonAvg of gIoU and cIoU: 78.61 -> 74.63Hurt
add CoT to baseline answer-side [SEG] (ablation)Baseline · EarthReason testimage attention mass: 0.246 -> 0.212Hurt
add Anticipatory Prior to baseline (ablation)GeoReason · EarthReason testimage attention mass: 0.212 -> 0.245 (answer-side), 0.270 prompt-sideImproved
rewrite queries as direct referring expressions (oracle)Baseline · EarthReasonAvg of gIoU and cIoU: 72.79 -> 80.84Improved

Later papers that compared with this method

1 later accepted paper reports GeoReason as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 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.

Experimental setup

Builds on: Embedding-as-Mask

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