# Finding Change in Satellite Archives from Text: How to Combine Before-and-After Images Efficiently

arXiv: https://arxiv.org/abs/2607.28571 · HTML: https://mlsanity.com/paper/2607.28571/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Point clouds, detection and remote sensing, one of 10 directions in State-space sequence models (S4, Mamba).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 30 Jul 2026.

## Placed in

- State-space sequence models (S4, Mamba) / [Point clouds, detection and remote sensing](https://mlsanity.com/ssm/point-clouds-detection-and-remote-sensing/)

## Builds on (most cited first, 12 of 12)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [SigLIP 2: Multilingual Vision-Language Encoders with Improved Semantic Understanding, Localization, and Dense Features](https://mlsanity.com/paper/2502.14786/) (2025)
- [RemoteCLIP: A Vision Language Foundation Model for Remote Sensing](https://mlsanity.com/paper/2306.11029/) (2023)
- [Fully Convolutional Siamese Networks for Change Detection](https://mlsanity.com/paper/1810.08462/) (2018)
- [RS5M and GeoRSCLIP: A Large Scale Vision-Language Dataset and A Large Vision-Language Model for Remote Sensing](https://mlsanity.com/paper/2306.11300/) (2023)
- [ChangeMamba: Remote Sensing Change Detection With Spatiotemporal State Space Model](https://mlsanity.com/paper/2404.03425/) (2024)
- [Cobra: Extending Mamba to Multi-Modal Large Language Model for Efficient Inference](https://mlsanity.com/paper/2403.14520/) (2024)
- [Relevance of Unsupervised Metrics in Task-Oriented Dialogue for Evaluating Natural Language Generation](https://mlsanity.com/paper/1706.09799/) (2017)
- [Towards a multimodal framework for remote sensing image change retrieval and captioning](https://mlsanity.com/paper/2406.13424/) (2024)
- [Self-Supervised Cross-Modal Text-Image Time Series Retrieval in Remote Sensing](https://mlsanity.com/paper/2501.19043/) (2025)

## Abstract (arXiv)

Operational Earth observation increasingly calls for answering queries such as ``find the image pairs where a new building appeared.'' This means searching an archive of before-and-after (bi-temporal) satellite image pairs and ranking each pair by how well it matches a natural-language description of the change. The component that performs this match, the fusion module that combines the ``before'' and ``after'' views, must be run at query time across many candidate pairs, so its speed largely sets the cost of every search. We present a controlled comparison of how to build that module. Using one fixed image encoder (a frozen CLIP model) and one training recipe for all variants, we evaluate eight designs drawn from three families: attention, state-space models (Mamba), and learned compression (our Temporal Bottleneck Fusion, TBF). Each design is tested on two benchmarks (LEVIR-CC and Dubai-CC) with ten random seeds, so the reported differences are statistically grounded. We outline three findings: first, a training-free two-stage search (a cheap difference model that shortlists candidates, followed by attention fusion that re-ranks them) matches or exceeds full-fusion recall on LEVIR-CC while cutting query cost 10-15×, with comparable R@1/R@5 on Dubai-CC; second, the linear-time scan of Mamba, attractive on paper, gives no speed benefit at the patch counts typical of vision transformers (L=196): the scan is limited by memory bandwidth, whereas attention maps cleanly onto parallel hardware; and third, compressing the fused representation (TBF) reduces parameters by 2.3× and latency by 1.6× for a change-only BLEU-1 cost of 0.007, although more aggressive compression quietly discards change-relevant detail that aggregate metrics fail to reveal.