AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Task-Oriented Data Synthesis and Control-Rectify Sampling for Remote Sensing Semantic Segmentation

Yunkai Yang, Yudong Zhang, Kunquan Zhang, Jinxiao Zhang, Xinying Chen, Haohuan Fu, Runmin Dong

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Retrieval and matching with generative features · 2 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add control-rectify flow matching (CRFM) samplingSD v3.5 Tri-attention · FUSU-4kmIoU: 48.57 -> 49.41Improved
add CRFM samplingSD v3.5 Tri-attention · LoveDA-5k (rural-to-urban)mIoU: 65.45 -> 66.12Improved
train on real plus TODSynth data vs real onlysegmentation baseline · FUSU-4kmIoU: 45.27 -> 49.41Improved
train on real plus TODSynth data vs real onlysegmentation baseline · LoveDA-5k (rural-to-urban)mIoU: 64.04 -> 66.12Improved
use Tri-attention instead of mask-adapter (ablation)SD v3.5 · FUSU-4kmIoU: 47.41 -> 48.57Improved
use Tri-attention instead of siamese MM-attention (ablation)SD v3.5 · FUSU-4kmIoU: 48.46 -> 48.57Improved
use FreeMask pixel-level filter instead of CLIP score filter (ablation)SD v3.5 · FUSU-4kmIoU: 46.69 -> 48.57Improved
increase CRFM steps from 0 to 2 (ablation)SD v3.5 (23 steps) · FUSU-4kmIoU: 48.57 -> 48.80No clear effect
increase CRFM steps from 4 to 6 (ablation)SD v3.5 (23 steps) · FUSU-4kFID: 38.65 -> 66.95Hurt
increase CRFM steps from 4 to 6 (ablation)SD v3.5 (23 steps) · FUSU-4kmIoU: 49.41 -> 48.74Hurt
apply CRFM with 6 steps at 18 sampling steps (ablation)SD v3.5 (18 steps) · FUSU-4kmIoU: 47.81 -> 49.57Mixed
optimise latent z_t directly with segmentation gradient instead of velocity field (ablation)CRFM · FUSUsynthetic data effectivenessHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy