AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Query-Kontext: An Unified Multimodal Model for Image Generation and Editing

YuXin Song, Wenkai Dong, Shizun Wang, Qi Zhang, Song Xue, Tao Yuan, Hu Yang, Haocheng Feng, Hang Zhou, Xinyan Xiao, Jingdong Wang

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Interleaved image-text generation with MLLMs

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 Query-Kontext instead of Qwen-Image on instruction editingQuery-Kontext · GEdit-Bench-ENG_O: 7.56 -> 7.66Improved
use Query-Kontext instead of Qwen-Image on instruction editingQuery-Kontext · GEdit-Bench-CNG_O: 7.52 -> 7.65Improved
use Query-Kontext instead of Qwen-Image for perceptual qualityQuery-Kontext · GEdit-Bench-ENG_PQ: 7.86 -> 7.37Hurt
use Query-Kontext instead of UNO-FLUX on DreamBoothQuery-Kontext · DreamBoothDINO: 0.760 -> 0.786Improved
use Query-Kontext instead of Metaquery on DreamBoothQuery-Kontext · DreamBoothCLIP-I: 0.851 -> 0.858Improved
use Query-Kontext instead of OmniGen on DreamBoothQuery-Kontext · DreamBoothCLIP-T: 0.315 -> 0.307Hurt
use Query-Kontext instead of OmniGen on DreamBenchQuery-Kontext · DreamBenchCLIP-T: 0.331 -> 0.336Improved
use Query-Kontext instead of UNO-FLUX on DreamBenchQuery-Kontext · DreamBenchDINO: 0.542 -> 0.532Hurt
use Query-Kontext instead of BAGEL on GenEvalQuery-Kontext · GenEvalOverall: 0.88 -> 0.88No clear effect
condition diffusion model on text plus Query-Kontext tokens from fine-tuned VLM instead of LLM text embeddings only (ablation)Query-Kontext Stage 2 · Diffusion alignmentconvergence speedImproved
increase LoRA rank across 64, 128, 256 (ablation)Query-Kontext · LoRA rank studyconvergence: faster convergence at higher ranks, marginal quality gains beyond r=128Mixed
unfreeze the diffusion model in Stage 2 (ablation)Query-Kontext · Stage 2 alignment (0.9B/4B/10B)alignment successImproved

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