Hoppa till innehÄllet
Global sound
Daily Paper Cast

Daily Paper Cast

WeLM-Image: A Unified Image Generation Model towards Visual Intelligence

· 27 min · Avsnitt 2486

Avsnittsbeskrivning

đŸ€— Upvotes: 28 | cs.CV, cs.AI

Authors:

Canmiao Fu, Chen Li, Guohui Zhang, Hao Lu, Hongyang Wei, Huadong Wu, Huaisong Zhang, Hubery Yin, Liyuan Cui, Qixin Yan, Tong Shao, Xiaoqiang Zhou, Xinhang Leng, Ying Zhang, Zhenhai Liu, Zhipeng Huang

Title:

WeLM-Image: A Unified Image Generation Model towards Visual Intelligence

Arxiv:

http://arxiv.org/abs/2611.07392v1

Abstract:

We present \textbf{WeLM-Image}, a unified image foundation model that integrates multimodal understanding into text-to-image generation, instruction-guided editing, and multi-reference synthesis. Designed for single-GPU deployment in WeChat's large-scale consumer applications, the model processes text, reference images, and target latents as one interleaved sequence. A Mixture-of-Transformers (MoT) backbone combines shared attention with independent understanding and generation parameters to limit optimization interference. Equipping each path with Mixture-of-Experts (MoE) modules provides sparse capacity, with 31.7B stored parameters and approximately 5.4B generation parameters activated per target latent token. A generation-oriented data engine combines a knowledge-grounded visual vocabulary with large-scale visual clustering to balance semantic coverage and editing tasks, while training-time evaluation guides iterative data refinement. Bilingual annotations and verified editing supervision support progressive training up to native 1.5K resolution. We optimize generation and editing separately with rubric- and checklist-based rewards, then consolidate the specialized policies through noise-aware on-policy distillation. In the reported overall comparisons, the 1536-resolution model ranks first among the evaluated open-source models on PRISM-Bench, Qwen-Image-Bench, and UnicBench. MeanFlow-based few-step distillation yields WeLM-Image-Turbo, which generates a $1024\times1024$ image in 4.98\,s on a single NVIDIA H20 GPU using eight guidance-free steps.

Webbplats: share.transistor.fm

Live, senaste timmen: 12 189 lÄtar pÄ 6 250 stationer i 142 lÀnder

UpptÀck vÀrlden genom radion

Fler sÀtt att upptÀcka (9)

Hört nÄgot bra? Spara det, eller öppna det direkt i Spotify.

★ Artist of the week The fastest riser on the radio this week
Olivia Rodrigo

expectations

Olivia Rodrigo · â–Č 6

Se videon Spotify

On the radio at Capital FM, Totally Radio Hits, Nova 10s, The Voice Radio

Your song here? From € 9 per week →
Kral TĂŒrk FMFortsĂ€tt lyssna