AI Paper Atlas

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

WISE: World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation

Yuwei Niu, Munan Ning, Mengren Zheng, Weiyang Jin, Bin Lin, Peng Jin, Jiaqi Liao, Chaoran Feng, Fanqing Meng, Kun-Peng Ning, Bin Zhu, Li Yuan

ICML 2026 · Image generation › Text-to-image diffusion · approach: T2I evaluation and controllable adapters

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
Rewrite WISE prompts into direct prompts with GPT-4o (ablation)20 T2I models · WISEWiScore overall: Janus-Pro-1B +0.34; FLUX.1-dev 0.50 -> 0.73Improved
Rewrite prompts for physics prompts on MetaQuery-XL (ablation)MetaQuery-XL · WISE PhysicsWiScore: -0.04Hurt
Add chain-of-thought before generation (ablation)BAGEL · WISEWiScore overall: 0.52 -> 0.70Improved
Use unified MLLM with AR+Diffusion architecture instead of dedicated T2IT2I models · WISEWiScore overall: Qwen-Image 0.62 vs FLUX.1-dev 0.50Improved
Use purely autoregressive unified modelJanus-Pro / Emu3 / VILA-U · WISEWiScore overall: Janus-1.3B 0.23, Janus-Pro-7B 0.35Hurt
Scale up AR model from Janus-Pro-1B to 7B (ablation)Janus-Pro · WISEWiScore overall: 0.26 -> 0.35Improved
Change WiScore weights to 0.5/0.3/0.2 or 0.4/0.4/0.2 (ablation)WiScore metric · WISE (gemini judge)model ranking: Spearman 0.993No clear effect
Swap GPT-4o judge for gemini-3.1-flash-lite or Qwen3.5-122B-A10B (ablation)WiScore evaluation · WISEmodel rankingNo clear effect
Vary random seed (42-46) (ablation)representative T2I models · WISEWiScore: Qwen-Image 0.5029 ± 0.0046No clear effect
Use recent AR models (NextFlow, Emu3.5) rather than early ARAR unified models · WISEWiScore: Emu3 0.39 -> NextFlow 0.62Improved
Rewrite prompts for physics (scientific categories) (ablation)T2I models · WISE Physicsrewrite-ineffective ratio: Physics 85.1%Mixed

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