# SARATR-X-v2: Scale-Aware Structural Pre-Training for SAR Foundation Models

arXiv: https://arxiv.org/abs/2607.23238 · HTML: https://mlsanity.com/paper/2607.23238/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 13 of its 118 references that are in the map.
- First posted on arXiv 25 Jul 2026.

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

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [VMamba: Visual State Space Model](https://mlsanity.com/paper/2401.10166/) (2024)
- [M4-SAR: A Multi-Resolution, Multi-Polarization, Multi-Scene, Multi-Source Dataset and Benchmark for optical-SAR Object Detection](https://mlsanity.com/paper/2505.10931/) (2025)
- [Visual Instruction Pretraining for Domain-Specific Foundation Models](https://mlsanity.com/paper/2509.17562/) (2025)
- [SARMAE: Masked Autoencoder for SAR Representation Learning](https://mlsanity.com/paper/2512.16635/) (2025)
- [A Complex-valued SAR Foundation Model Based on Physically Inspired Representation Learning](https://mlsanity.com/paper/2504.11999/) (2025)
- [CrossEarth-SAR: A SAR-Centric and Billion-Scale Geospatial Foundation Model for Domain Generalizable Semantic Segmentation](https://mlsanity.com/paper/2603.12008/) (2026)
- [On the Status of Foundation Models for SAR Imagery](https://mlsanity.com/paper/2509.21722/) (2025)
- [FUSAR-KLIP: Towards Multimodal Foundation Models for Remote Sensing](https://mlsanity.com/paper/2509.23927/) (2025)
- [HuiYanEarth-SAR: A Foundation Model for High-Fidelity and Low-Cost Global Remote Sensing Imagery Generation](https://mlsanity.com/paper/2604.11444/) (2026)
- [SARVLM: A Vision Language Foundation Model for Semantic Understanding in SAR Imagery](https://mlsanity.com/paper/2510.22665/) (2025)
- [SAMBA: A Scatter-Guided Masked Bidirectional Mamba Foundation Model for SAR Target Recognition](https://mlsanity.com/paper/2606.31668/) (2026)

## Abstract (arXiv)

Masked image modeling has become a dominant paradigm for SAR pre-training, yet the design of the reconstruction target remains fundamentally unsettled. This article argues that a SAR pre-training target should satisfy two conditions to produce transferable representations: (i) physics-grounded stability, i.e., approximate invariance of the target operator to multiplicative speckle inherent in coherent imaging; and (ii) semantic scale compatibility, i.e., coverage of the heterogeneous spatial scales that downstream tasks demand. These two conditions are individually achievable but jointly difficult: physics-grounded stability favors fixed operators, while semantic scale compatibility favors data-driven composition. To this end, SARATR-X-v2 reconciles both within a single design. The target is constructed through fixed structural extractors spanning six receptive fields, from blind-spot local aggregation to directional log-ratio region contrast, and fused via learnable weights into one unified supervision signal for masked reconstruction. On twelve SAR benchmarks across classification, detection, and segmentation, SARATR-X-v2 achieves state-of-the-art transfer performance. Under synthetic speckle variation, the proposed target reduces perturbation drift in the learned supervision by nearly two orders of magnitude relative to pixel-space supervision. Taken together, these results support physics-grounded stability and semantic scale compatibility as a principled framework for pre-training target design under coherent imaging, and suggest that effective SAR pre-training is not about reconstructing more signal, but about reconstructing the right structural target.