# FM-ChangeNet: Learning Change through Pathwise Feature Transport

arXiv: https://arxiv.org/abs/2607.04750 · HTML: https://mlsanity.com/paper/2607.04750/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 18 of its 35 references that are in the map.
- First posted on arXiv 6 Jul 2026.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 18 of 18)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Improving and generalizing flow-based generative models with minibatch optimal transport](https://mlsanity.com/paper/2302.00482/) (2023)
- [Fully Convolutional Siamese Networks for Change Detection](https://mlsanity.com/paper/1810.08462/) (2018)
- [A Transformer-Based Siamese Network for Change Detection](https://mlsanity.com/paper/2201.01293/) (2022)
- [ChangeMamba: Remote Sensing Change Detection With Spatiotemporal State Space Model](https://mlsanity.com/paper/2404.03425/) (2024)
- [RSMamba: Remote Sensing Image Classification with State Space Model](https://mlsanity.com/paper/2403.19654/) (2024)
- [S2Looking: A Satellite Side-Looking Dataset for Building Change Detection](https://mlsanity.com/paper/2107.09244/) (2021)
- [The Multi-Temporal Urban Development SpaceNet Dataset](https://mlsanity.com/paper/2102.04420/) (2021)
- [CDMamba: Incorporating Local Clues into Mamba for Remote Sensing Image Binary Change Detection](https://mlsanity.com/paper/2406.04207/) (2024)
- [Fully Convolutional Change Detection Framework with Generative Adversarial Network for Unsupervised, Weakly Supervised and Regional Supervised Change Detection](https://mlsanity.com/paper/2201.06030/) (2022)
- [Super-resolution-based Change Detection Network with Stacked Attention Module for Images with Different Resolutions](https://mlsanity.com/paper/2103.00188/) (2021)
- [Continuous Cross-resolution Remote Sensing Image Change Detection](https://mlsanity.com/paper/2305.14722/) (2023)
- [ChangeAnywhere: Sample Generation for Remote Sensing Change Detection via Semantic Latent Diffusion Model](https://mlsanity.com/paper/2404.08892/) (2024)
- [UCDFormer: Unsupervised Change Detection Using a Transformer-driven Image Translation](https://mlsanity.com/paper/2308.01146/) (2023)
- [NeXt2Former-CD: Efficient Remote Sensing Change Detection with Modern Vision Architectures](https://mlsanity.com/paper/2602.18717/) (2026)
- [DiffRegCD: Integrated Registration and Change Detection with Diffusion Features](https://mlsanity.com/paper/2511.07935/) (2025)

## Abstract (arXiv)

We present FM-ChangeNet, a pathwise-supervised framework for change detection that reformulates bi-temporal reasoning as continuous transport in feature space rather than static endpoint comparison. Given encoded pre and post-temporal representations, we construct intermediate latent states and learn a time-conditioned velocity field hatvθ(zt,t) along the transformation trajectory. This pathwise formulation constrains the predictor over a continuum of intermediate states, providing a denser and less ambiguous supervision signal than conventional endpoint-only segmentation and enabling the model to capture temporal evolution explicitly. The learned velocity field is not only a transport mechanism but also an interpretable representation of change: its magnitude serves as a spatially localized change cue that helps distinguish true structural variation from nuisance effects such as illumination shifts and spatial misalignment. We develop a hierarchical multi-scale architecture with cross-temporal alignment, time-conditioned coarse-to-fine flow decoding, and a unified objective that couples flow supervision, trajectory consistency, spatial regularization, and segmentation loss. Experiments on remote sensing benchmarks show that the proposed framework produces more structured and robust change representations while achieving state-of-the-art performance.