AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Visual Programming for Step-by-Step Text-to-Image Generation and Evaluation

Jaemin Cho, Abhay Zala, Mohit Bansal

NeurIPS 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Vision-language pretraining · 4 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
train layout LM on Flickr30K+COCO+PaintSkills instead of only Flickr30K (ablation)VPGEN · skill-based promptsaverage skill score: 43.9 -> 51.0Improved
add COCO and PaintSkills to layout training (ablation)VPGEN · open-ended promptsscore: 71.0 -> 68.3Hurt
add COCO and PaintSkills to layout training (ablation)VPGEN · COCO 30KFID: 15.9 -> 20.1Hurt
step-by-step layout generation instead of direct generationVPGEN vs Stable Diffusion v1.4 · Count skillVPEval accuracy: 47.4 -> 72.2Improved
step-by-step layout generation instead of direct generationVPGEN vs Stable Diffusion v1.4 · Spatial skillVPEval accuracy: 22.9 -> 56.1Improved
step-by-step layout generation instead of direct generationVPGEN vs Stable Diffusion v1.4 · Text Rendering skillVPEval accuracy: 8.9 -> 3.7Hurt
train on synthetic PaintSkills only instead of real Flickr30K (ablation)VPGEN · skill-based / open-endedaverage score: 43.9 -> 46.0 (skill); 71.0 -> 65.6 (open-ended)Mixed
use 1000 bins instead of 100 for box quantization (ablation)VPGEN · skill-based promptsaverage skill score: 51.0 -> 49.8No clear effect
use 1000 bins instead of 100 (ablation)VPGEN · open-ended promptsscore: 68.3 -> 68.2No clear effect
replace finetuned Vicuna 13B with GPT-3.5-Turbo (36 in-context examples) (ablation)VPGEN · skill-based promptsaverage skill score: 51.0 -> 40.7Hurt
replace finetuned Vicuna 13B with GPT-3.5-Turbo (ablation)VPGEN · open-ended promptsscore: 68.3 -> 65.5Hurt
use BLIP-2 VQA for object/spatial/scale modules in VPEval (ablation)VPEval · skill-based human correlationSpearman rho overall: 73.5 -> 79.0Improved

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