mlsanityData to 7 Oct 2026

GeoRoPE: Ground-Aware Rotary Adaptation for Remote Sensing Foundation Models

arXiv 2606.14760 ↗First posted 8 Jun 2026cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

Remote-sensing foundation models (RSFMs) benefit from pretraining on imagery from multiple sensors and ground sampling distances (GSDs), but such exposure alone does not resolve scale mismatch during downstream adaptation. A fixed token-grid offset can correspond to different ground distances across sensors, making grid-based positional priors physically inconsistent. Meanwhile, heterogeneous spatial granularity means that compact urban regions and homogeneous landscapes may require different positional sensitivities even under the same GSD. Therefore, we propose GeoRoPE, a ground-aware, RoPE-compatible, and parameter-efficient spatial adaptation method for RSFMs. GeoRoPE recalibrates token-level positional interactions from two complementary aspects. First, textitGeo-Coordinate Calibration (GCC) rescales raw token-grid offsets according to the ground distance represented by one token-grid step, producing geo-calibrated relative coordinates across GSDs. Second, textitGeo-Frequency Calibration (GFC) adjusts the native RoPE frequency with a relation-specific factor, enabling position sensitive adaptation to scene-dependent spatial granularity. GeoRoPE is injected into pretrained RSFMs through a lightweight adapter, preserving the frozen spatial prior while adding geo-aware positional corrections. Experiments across multiple RSFMs, sensors, resolutions, and downstream tasks demonstrate that GeoRoPE improves cross-resolution robustness and scale-sensitive representation learning.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 32 of its 37 references that are in the map. First posted on arXiv 8 Jun 2026.

Builds on

32 of 37 references.

References · 32

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  3. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  4. 2018Unified Perceptual Parsing for Scene Understandingcited by 1,265
  5. 2021BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-modelscited by 1,234
  6. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,215
  7. 2023YaRN: Efficient Context Window Extension of Large Language Modelscited by 584
  8. 2022SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagerycited by 402
  9. 2022Scale-MAE: A Scale-Aware Masked Autoencoder for Multiscale Geospatial Representation Learningcited by 302
  10. 2024LongRoPE: Extending LLM Context Window Beyond 2 Million Tokenscited by 302
  11. 2023SpectralGPT: Spectral Remote Sensing Foundation Modelcited by 250
  12. 2024Rotary Position Embedding for Vision Transformercited by 226
20 more
  1. 2021Rethinking and Improving Relative Position Encoding for Vision Transformercited by 206
  2. 2023SkySense: A Multi-Modal Remote Sensing Foundation Model Towards Universal Interpretation for Earth Observation Imagerycited by 202
  3. 2022SatlasPretrain: A Large-Scale Dataset for Remote Sensing Image Understandingcited by 164
  4. 2023GEO-Bench: Toward Foundation Models for Earth Monitoringcited by 122
  5. 2023PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Trainingcited by 109
  6. 2018Urban Change Detection for Multispectral Earth Observation Using Convolutional Neural Networkscited by 108
  7. 2024Neural Plasticity-Inspired Multimodal Foundation Model for Earth Observationcited by 97
  8. 2025Cameras as Relative Positional Encodingcited by 94
  9. 2024AnySat: One Earth Observation Model for Many Resolutions, Scales, and Modalitiescited by 73
  10. 2023Tuning LayerNorm in Attention: Towards Efficient Multi-Modal LLM Finetuningcited by 63
  11. 2024OmniSat: Self-Supervised Modality Fusion for Earth Observationcited by 63
  12. 2025Towards a Unified Copernicus Foundation Model for Earth Visioncited by 52
  13. 2024Cross-Scale MAE: A Tale of Multi-Scale Exploitation in Remote Sensingcited by 47
  14. 2024Bridging Remote Sensors with Multisensor Geospatial Foundation Modelscited by 46
  15. 2024One for All: Toward Unified Foundation Models for Earth Visioncited by 34
  16. 2023Multi-modal learning for geospatial vegetation forecastingcited by 16
  17. 2024An Efficient Recipe for Long Context Extension via Middle-Focused Positional Encodingcited by 16
  18. 2025Parameter-Efficient Adaptation of Geospatial Foundation Models through Embedding Deflectioncited by 10
  19. 2025SpectralX: Parameter-efficient Domain Generalization for Spectral Remote Sensing Foundation Modelscited by 5
  20. 2025PIPE: Physics-Informed Position Encoding for Alignment of Satellite Images and Time Seriescited by 1

Built on by

Not cited yet.