AI Paper Atlas

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

Are We Ready for RL in Text-to-3D Generation? A Progressive Investigation

Yiwen Tang, Zoey Guo, Kaixin Zhu, Ray Zhang, Qizhi Chen, Dongzhi Jiang, Junli Liu, Bohan Zeng, Haoming Song, Delin Qu, Tianyi Bai, Dan Xu and 2 more

CVPR 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters · 11 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 textual reasoning before 3D token generation in GRPO (ablation)ShapeLLM-Omni + GRPO (HPS reward) · Toys4K (800 sampled)CLIP score: 23.4 -> 24.0Improved
add HPS v2.1 human preference reward (ablation)ShapeLLM-Omni + GRPO · Toys4K (800 sampled)CLIP score: 22.7 -> 24.0Improved
add UnifiedReward on top of HPS (ablation)GRPO reward · Toys4K (800 sampled)CLIP score: 24.0 -> 24.6Improved
replace UnifiedReward with Qwen2.5-VL for alignment next to HPS (ablation)GRPO reward · Toys4K (800 sampled)CLIP score: 24.6 -> 24.2Hurt
add Qwen2.5-VL 3D consistency reward to HPS + UnifiedReward (ablation)GRPO reward · Toys4K (800 sampled)CLIP score: 24.6 -> 25.2Improved
add dynamic sampling to GRPO (ablation)GRPO · Toys4K (800 sampled)CLIP score: 25.2 -> 25.8Improved
add token-level loss averaging on top of dynamic sampling (ablation)GRPO + dynamic sampling · Toys4K (800 sampled)CLIP score: 25.8 -> 26.3Improved
add sequence-level optimization (GSPO) on top of dynamic sampling (ablation)GRPO + dynamic sampling · Toys4K (800 sampled)CLIP score: 25.8 -> 25.5Hurt
remove KL penalty (ablation)GRPO + dynamic sampling + token averaging · Toys4K (800 sampled)CLIP score: 26.3 -> 25.9Hurt
add decoupled clip to sampling + token averaging (ablation)GRPO · Toys4K (800 sampled)CLIP score: 26.3 -> 26.5Improved
scale training data 1.5x, 2x, 3x (ablation)best RL configuration · Toys4K (800 sampled)CLIP score: +0.4, +0.2, +0.4Improved
double training iterations (ablation)best RL configuration · Toys4K (800 sampled)CLIP score: +0.9Improved

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