AI Paper Atlas

Research map / Image generation / Text-to-image diffusion

SpatialReward: Verifiable Spatial Reward Modeling for Fine-Grained Spatial Consistency in Text-to-Image Generation

Sashuai Zhou, Qiang Zhou, Junpeng Ma, Yue Cao, Ruofan Hu, Ziang Zhang, Xiaoda Yang, Zhibin Wang, Jun Song, Cheng Yu, Bo Zheng, Zhou Zhao

CVPR 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 6 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
use SpatialReward as RL rewardSD3.5-M with Flow-GRPO · GenEval (80-Obj)overall score: 0.67 -> 0.95Improved
use SpatialReward as RL rewardSD3.5-M with Flow-GRPO · SpatRelBench (1k-Obj)overall score: 0.23 -> 0.42Improved
use SpatialReward as RL rewardFLUX1-dev with Flow-GRPO · GenEval (80-Obj)overall score: 0.76 -> 0.97Improved
use SpatialReward as RL rewardFLUX1-dev with Flow-GRPO · SpatRelBench (1k-Obj)overall score: 0.28 -> 0.46Improved
use SpatialReward instead of UnifiedRewardSD3.5-M with Flow-GRPO · SpatRelBench (1k-Obj)overall score: 0.33 -> 0.42Improved
use QwenVL as the RL rewardSD3.5-M with Flow-GRPO · GenEval (80-Obj)overall score: 0.67 -> 0.61Hurt
add exclusion constraints (ablation)SpatialReward · SpatRelBenchaccuracy: 25.9 -> 37.1Improved
add expert detection (ablation)SpatialReward · GenEvalaccuracy: 70.3 -> 95.2Improved
add expert detection (ablation)SpatialReward · SpatRelBenchaccuracy: 21.6 -> 37.1Improved
add chain-of-thought reasoning (ablation)SpatialReward · SpatRelBenchaccuracy: 27.9 -> 37.1Improved
add chain-of-thought reasoning (ablation)SpatialReward · GenEvalaccuracy: 94.2 -> 95.2Improved
train SD3.5-M with SpatialRewardSD3.5-M · Aestheticaesthetic score: 5.39 -> 5.23Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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