AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DiffusionVL: Translating Any Autoregressive Models into Diffusion Vision Language Models

Lunbin Zeng, Jingfeng Yao, Bencheng Liao, Hongyuan Tao, Wenyu Liu, Xinggang Wang

ECCV 2026 · Image generation › Fast diffusion sampling · approach: Masked diffusion language model decoding · 16 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
use training block size 1 instead of 8 (ablation)DiffusionVL-3B · ChartQAaccuracy: 79.9 -> 82.8Improved
use training block size 16 instead of 8 (ablation)DiffusionVL-3B · ChartQAaccuracy: 79.9 -> 77.4Hurt
increase training block size from 1 to 16 (ablation)DiffusionVL-3B · DetailCaps throughputTPS: 26.9 -> 41.6Improved
use block size 8 rather than 1 or 16 (ablation)DiffusionVL-3B · DetailCapsBERTScore: 29.8 -> 30.9Improved
add block-level noise instead of sequence-level noise (ablation)DiffusionVL-7B · MMMU-Pro (vis.)accuracy: 24.86 -> 24.97Improved
add block-level noise instead of sequence-level noise (ablation)DiffusionVL-7B · MME (perp.)score: 1527 -> 1519Hurt
use block diffusion instead of full diffusion (ablation)DiffusionVL-7B · ChartQAaccuracy: 52.00 -> 84.20Improved
use block diffusion instead of full diffusion (ablation)DiffusionVL-7B · MME (perp.)score: 1607 -> 1519Hurt
lower dynamic remasking threshold from 1.0 to 0.2 (ablation)DiffusionVL-3B · DetailCapsTPS and BERTScore: TPS 31.5 -> 71.8; BERT 30.2 -> 18.6Mixed
lower dynamic remasking threshold from 1.0 to 0.6 (ablation)DiffusionVL-3B · DetailCapsTPS and BERTScore: TPS 31.5 -> 37.8; BERT 30.2 -> 29.7Mixed
start from AR-LM Qwen2.5-7B instead of dLLM LLaDA-8B for block diffusion VLMDiffusionVL · ChartQAaccuracy: 29.8 -> 53.6Improved
use block diffusion instead of AR finetuning on the same AR-LMDiffusionVL (Qwen2.5-7B) · MMMU valaccuracy: 45.4 -> 43.7No clear effect

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