RS-WorldModel: a Unified Model for Remote Sensing Understanding and Future Sense Forecasting

TL;DR

RS-WorldModel is a unified remote sensing model that jointly understands changes and forecasts future scenes using a large annotated dataset, outperforming existing models in accuracy and efficiency.

Contribution

The paper introduces RS-WorldModel, a novel unified model for remote sensing that combines change understanding and future scene forecasting, along with a large dataset RSWBench-1.1M.

Findings

Outperforms larger open-source models in spatiotemporal question-answering.

Achieves an FID of 43.13 on text-guided scene forecasting.

Uses only 2B parameters, demonstrating efficiency.

Abstract

Remote sensing world models aim to both explain observed changes and forecast plausible futures, two tasks that share spatiotemporal priors. Existing methods, however, typically address them separately, limiting cross-task transfer. We present RS-WorldModel, a unified world model for remote sensing that jointly handles spatiotemporal change understanding and text-guided future scene forecasting, and we build RSWBench-1.1M, a 1.1 million sample dataset with rich language annotations covering both tasks. RS-WorldModel is trained in three stages: (1) Geo-Aware Generative Pre-training (GAGP) conditions forecasting on geographic and acquisition metadata; (2) synergistic instruction tuning (SIT) jointly trains understanding and forecasting; (3) verifiable reinforcement optimization (VRO) refines outputs with verifiable, task-specific rewards. With only 2B parameters, RS-WorldModel surpasses open-source models up to 120$ \times $ larger on most spatiotemporal change question-answering metrics. It achieves an FID of 43.13 on text-guided future scene forecasting, outperforming all open-source baselines as well as the closed-source Gemini-2.5-Flash Image (Nano Banana).