AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Render-in-the-Loop: Vector Graphics Generation via Visual Self-Feedback

Guotao Liang, Zhangcheng Wang, Juncheng Hu, Haitao Zhou, Ziteng Xue, Jing Zhang, Dong Xu, Qian Yu

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: VLM capability analysis · 20 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 Visual Self-Feedback training (ablation)Qwen3-VL-8B SFT · MMSVG-IconFID: 148.72 -> 127.98Improved
add Visual Self-Feedback training (ablation)Qwen3-VL-8B SFT · MMSVG-IllustrationFID: 155.84 -> 137.86Improved
add Visual Self-Feedback training (ablation)Qwen3-VL-8B SFT · MMSVG-IconLPIPS: 0.312 -> 0.188Improved
add Render-and-Verify at inference (ablation)Ours (VSF) · MMSVG-IconFID: 127.98 -> 127.64Improved
add Render-and-Verify at inference (ablation)Ours (VSF) · MMSVG-IconLPIPS: 0.188 -> 0.172Improved
add Render-and-Verify at inference (ablation)Ours (VSF) · MMSVG-IllustrationLPIPS: 0.193 -> 0.178Improved
add Render-and-Verify at inference (ablation)Ours (VSF) · MMSVG-IllustrationFID: 137.86 -> 137.79No clear effect
add naive multi-turn visual feedback without training (ablation)GPT-5 · MMSVG-IllustrationLPIPS: 0.345 -> 0.388Hurt
add naive multi-turn visual feedback without training (ablation)GPT-5 · MMSVG-IconLPIPS: 0.387 -> 0.405Hurt
add naive multi-turn visual feedback without training (ablation)Qwen3-VL / GPT-5 · MMSVG-IconCLIP: 0.311 -> 0.294Hurt
replace OmniSVG with Render-in-the-Loop (0.85M vs 2M data)Ours (VSF+RaV) · MMSVG-IconFID: 130.56 -> 127.64Improved
replace InternSVG (16M data) with Render-in-the-LoopOurs (VSF+RaV) · MMSVG-IconFID: 128.80 -> 127.64Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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