AI Paper Atlas

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

Enhancing Spatial Understanding in Image Generation via Reward Modeling

Zhenyu Tang, Chaoran Feng, Yufan Deng, Jie Wu, Xiaojie Li, Rui Wang, Yunpeng Chen, Daquan Zhou

CVPR 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 3 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
scale SpatialScore backbone from Qwen2.5-VL-3B to 7B (ablation)SpatialScore · Spatial reward benchmark (365 pairs)pairwise accuracy: 0.891 -> 0.958Improved
scale SpatialScore backbone from 7B to 32B (ablation)SpatialScore · Spatial reward benchmark (365 pairs)pairwise accuracy: 0.958 -> 0.973Improved
add top-k filtering (k=6) to GRPO (ablation)Flux.1-dev GRPO · SpatialScore in-domainSpatialScore reward: 7.73 -> 7.81Improved
add top-k filtering (k=6) to GRPO (ablation)Flux.1-dev GRPO · DPG-Bench Relation-Spatialscore: 0.919 -> 0.932Improved
add top-k filtering (k=6) to GRPO (ablation)Flux.1-dev GRPO · UniGenBench++ long Layout-2Dscore: 0.891 -> 0.887Hurt
add top-k filtering (k=6) to GRPO (ablation)Flux.1-dev GRPO · UniGenBench++ long Layout-3Dscore: 0.793 -> 0.801Improved
add top-k filtering (k=6) to GRPO (ablation)Flux.1-dev GRPO · policy-update NFE per promptNFE: 24*6 -> 12*6Improved
use k=4 instead of k=6 in top-k filtering (ablation)Flux.1-dev GRPO · SpatialScore in-domainSpatialScore reward: 7.81 -> 7.71Hurt
use k=4 instead of k=6 in top-k filtering (ablation)Flux.1-dev GRPO · DPG-Bench Relation-Spatialscore: 0.932 -> 0.916Hurt
train Flux.1-dev with SpatialScore RL instead of GenEval reward (Flow-GRPO)Flux.1-dev · DPG-Bench overalloverall score: 57.02 -> 85.03Improved
train Flux.1-dev with SpatialScore RLFlux.1-dev · DPG-Bench overalloverall score: 82.91 -> 85.03Improved
train Flux.1-dev with SpatialScore RLFlux.1-dev · SpatialScore in-domainSpatialScore reward: 2.18 -> 7.81Improved

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