AI Paper Atlas

Research map / Vision-language and multimodal / Robot action and navigation

WeMMU: Enhanced Bridging of Vision-Language Models and Diffusion Models via Noisy Query Tokens

Jian Yang, Dacheng Yin, Xiaoxuan He, Yong Li, Fengyun Rao, Jing Lyu, Wei Zhai, Yang Cao, Zheng-Jun Zha

CVPR 2026 · Vision-language and multimodal › Robot action and navigation · approach: Embodied multimodal LLM agents · 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
make learnable queries dynamic by resolution (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · ImageEdit-Bench (Stage 3)Overall: 2.53 -> 2.88Improved
replace learnable queries with Noisy Query Tokens (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · ImageEdit-Bench (Stage 3)Overall: 2.88 -> 2.98Improved
add VAE branch with linear layer (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · ImageEdit-Bench (Stage 3)Overall: 2.98 -> 3.31Improved
add noisy queries vs learnable queries for multi-image editing (Stage 4) (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · multi-image editingqualitative successImproved
add learnable channel-wise scaling to noise (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · WeMMUperformance: scale mean 1.0, std 0.0074No clear effect
add Stage 4 multi-image trainingWeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · GenEvaloverall: 0.88 -> 0.88No clear effect
add Stage 4 multi-image trainingWeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · ImageEdit-BenchOverall: 3.31 -> 3.30No clear effect
unfreeze Qwen2.5-VL ViT (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · WeMMUtraining stabilityHurt
connect VAE branch with a linear layer instead of 2/6-layer ViT (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · WeMMUtraining loss convergenceImproved
use contrastive flow matching in later stages (ablation)WeMMU query bridge (Qwen2.5-VL-3B + Sana 1.6B) · WeMMUperformanceNo clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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