AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Beyond Atomic Layouts: Compositional Design Understanding with Vision-Language Models

Yiyang Huang, Zhaowen Wang, Simon Jenni, Jing Shi, Yitian Zhang, Yizhou Wang, Yun Fu

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks · 0 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 multimodal alignment (grounding) to direct finetuning (ablation)Direct Finetune on Qwen2.5-VL 7B · CoDeLayout (type-balanced)weighted accuracy: 80.21 -> 82.03Improved
add structural perception (spatial relationship metadata) to direct finetuning (ablation)Direct Finetune on Qwen2.5-VL 7B · CoDeLayout (type-balanced)weighted accuracy: 80.21 -> 83.85Improved
combine MA and SP (full MASON) versus either alone (ablation)Qwen2.5-VL 7B · CoDeLayout (type-balanced)weighted accuracy: 83.85 -> 84.11Improved
add MA on Morphing category (ablation)Direct Finetune · CoDeLayout Morphing (type-balanced)accuracy: 72.34 -> 70.21Hurt
train MASON on 30% type-balanced data instead of Direct Finetune on full dataQwen2.5-VL 7B · CoDeLayout testweighted accuracy: 88.80 -> 89.32Improved
MASON versus Direct Finetune across data scales 10-100% (ablation)Qwen2.5-VL 7B · CoDeLayout testweighted accuracyImproved
replace GPT-4o grounding generator with Qwen3-VL script-based grounding (ablation)MASON · CoDeLayout testweighted accuracy: 91.6 -> 91.1No clear effect
remove visual input at test time (ablation)MASON · CoDeLayout testweighted accuracy: 91.6 -> 65.1Hurt
fine-tune MASON on full data versus 30% balanced data on MorphingMASON · CoDeLayout Morphingaccuracy: 72.34 -> 70.21Hurt
post-train Qwen2.5-VL 7B with MASON versus zero-shot GPT-o3Qwen2.5-VL 7B · CoDeLayout Clippingaccuracy: 63.64 -> 90.91Improved

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