# One Query, Many Scales: Sparse Mixture-of-Experts for Efficient Hierarchical Cross-View Geo-Localization

arXiv: https://arxiv.org/abs/2608.01060 · HTML: https://mlsanity.com/paper/2608.01060/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts.
- Builds on 13 of its 68 references that are in the map.
- First posted on arXiv 2 Aug 2026.

## Placed in

- Mixture-of-experts / [Vision mixture-of-experts](https://mlsanity.com/moe/vision-mixture-of-experts/)

## Builds on (most cited first, 13 of 13)

- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Layer Normalization](https://mlsanity.com/paper/1607.06450/) (2016)
- [Gaussian Error Linear Units (GELUs)](https://mlsanity.com/paper/1606.08415/) (2016)
- [Echo-4o: Harnessing the Power of GPT-4o Synthetic Images for Improved Image Generation](https://mlsanity.com/paper/2508.09987/) (2025)
- [Simple, Effective and General: A New Backbone for Cross-view Image Geo-localization](https://mlsanity.com/paper/2302.01572/) (2023)
- [CrossViewDiff: A Cross-View Diffusion Model for Satellite-to-Street View Synthesis](https://mlsanity.com/paper/2408.14765/) (2024)
- [Mind-Brush: Integrating Agentic Cognitive Search and Reasoning into Image Generation](https://mlsanity.com/paper/2602.01756/) (2026)
- [RealGen: Photorealistic Text-to-Image Generation via Detector-Guided Rewards](https://mlsanity.com/paper/2512.00473/) (2025)
- [GenClaw: Code-Driven Agentic Image Generation](https://mlsanity.com/paper/2605.30248/) (2026)
- [A Unified Hierarchical Framework for Fine-grained Cross-view Geo-localization over Large-scale Scenarios](https://mlsanity.com/paper/2505.07622/) (2025)
- [SMGeo: Cross-View Object Geo-Localization with Grid-Level Mixture-of-Experts](https://mlsanity.com/paper/2511.14093/) (2025)
- [Learnable Query Aggregation with KV Routing for Cross-view Geo-localisation](https://mlsanity.com/paper/2512.23938/) (2025)
- [Just Zoom In: Cross-View Geo-Localization via Autoregressive Zooming](https://mlsanity.com/paper/2603.25686/) (2026)

## Abstract (arXiv)

Cross-view geo-localization (CVGL) retrieves geo-tagged satellite imagery for a ground-view query. Most systems exhaustively search a flat, fixed-resolution gallery, incurring high cost over large areas and adapting poorly to satellite resolution changes. Autoregressive coarse-to-fine alternatives reduce comparisons but bind later predictions to earlier decisions and a predefined hierarchy. We introduce GeoMoE, a sparse mixture-of-experts dual encoder that decouples global multi-scale representation learning from local hierarchical search. Global multi-scale supervision and content-adaptive routing map ground and satellite images across resolutions into a globally comparable embedding space. At inference, each image is encoded once, and probabilistic beam search follows parent--child links to score a small candidate subset. Later levels reuse these descriptors rather than features generated by preceding levels, limiting feature-level error propagation and hierarchy coupling. We further introduce VIGOR-M, a four-city benchmark with an explicit parent--child satellite hierarchy and held-out half-step galleries for single-resolution, cross-resolution, and hierarchical evaluation. GeoMoE achieves 95.78% R@40m on Just Zoom In, 2.77 percentage points above the previous best, and 62.39% R@1 on VIGOR-M. The latter requires 0.885 MMAC/query for descriptor matching, 5.27% of an exhaustive L3 scan, while exceeding the strongest exhaustive baseline by 3.12 percentage points in R@1. One model trained on L1, L2, and L3 also outperforms a matched dense control across all six galleries and transfers to three withheld resolutions. By decoupling globally trained embeddings from local hierarchical search, GeoMoE jointly improves localization accuracy, search efficiency, and cross-resolution transfer.