Salvato in:
Dettagli Bibliografici
Autori principali: Wang, Peiyu, Peng, Yi, Gan, Yimeng, Hu, Liang, Xie, Tianyidan, Wang, Xiaokun, Wei, Yichen, Tang, Chuanxin, Zhu, Bo, Li, Changshi, Wei, Hongyang, Li, Eric, Song, Xuchen, Liu, Yang, Zhou, Yahui
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2508.03320
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911093115322368
author Wang, Peiyu
Peng, Yi
Gan, Yimeng
Hu, Liang
Xie, Tianyidan
Wang, Xiaokun
Wei, Yichen
Tang, Chuanxin
Zhu, Bo
Li, Changshi
Wei, Hongyang
Li, Eric
Song, Xuchen
Liu, Yang
Zhou, Yahui
author_facet Wang, Peiyu
Peng, Yi
Gan, Yimeng
Hu, Liang
Xie, Tianyidan
Wang, Xiaokun
Wei, Yichen
Tang, Chuanxin
Zhu, Bo
Li, Changshi
Wei, Hongyang
Li, Eric
Song, Xuchen
Liu, Yang
Zhou, Yahui
contents We introduce Skywork UniPic, a 1.5 billion-parameter autoregressive model that unifies image understanding, text-to-image generation, and image editing within a single architecture-eliminating the need for task-specific adapters or inter-module connectors-and demonstrate that compact multimodal systems can achieve state-of-the-art performance on commodity hardware. Skywork UniPic achieves a GenEval score of 0.86, surpassing most existing unified models; sets a new DPG-Bench complex-generation record of 85.5; attains 5.83 on GEditBench-EN and 3.49 on ImgEdit-Bench for image editing; and generates 1024 x 1024 images with under 15 GB of GPU memory (e.g., RTX 4090). (1) a decoupled encoding strategy that leverages a masked autoregressive encoder for synthesis and a SigLIP2 encoder for understanding, all feeding a shared autoregressive decoder; (2) a progressive, resolution-aware training schedule scaling from 256 x 256 to 1024 x 1024 while dynamically unfreezing parameters to balance capacity and stability; and (3) meticulously curated, 100 million-scale datasets augmented with task-specific reward models to refine generation and editing objectives. By demonstrating that high-fidelity multimodal integration need not incur prohibitive resource demands, Skywork UniPic establishes a practical paradigm for deployable, high-fidelity multimodal AI. Code and weights are publicly available at https://huggingface.co/Skywork/Skywork-UniPic-1.5B.
format Preprint
id arxiv_https___arxiv_org_abs_2508_03320
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation
Wang, Peiyu
Peng, Yi
Gan, Yimeng
Hu, Liang
Xie, Tianyidan
Wang, Xiaokun
Wei, Yichen
Tang, Chuanxin
Zhu, Bo
Li, Changshi
Wei, Hongyang
Li, Eric
Song, Xuchen
Liu, Yang
Zhou, Yahui
Computer Vision and Pattern Recognition
We introduce Skywork UniPic, a 1.5 billion-parameter autoregressive model that unifies image understanding, text-to-image generation, and image editing within a single architecture-eliminating the need for task-specific adapters or inter-module connectors-and demonstrate that compact multimodal systems can achieve state-of-the-art performance on commodity hardware. Skywork UniPic achieves a GenEval score of 0.86, surpassing most existing unified models; sets a new DPG-Bench complex-generation record of 85.5; attains 5.83 on GEditBench-EN and 3.49 on ImgEdit-Bench for image editing; and generates 1024 x 1024 images with under 15 GB of GPU memory (e.g., RTX 4090). (1) a decoupled encoding strategy that leverages a masked autoregressive encoder for synthesis and a SigLIP2 encoder for understanding, all feeding a shared autoregressive decoder; (2) a progressive, resolution-aware training schedule scaling from 256 x 256 to 1024 x 1024 while dynamically unfreezing parameters to balance capacity and stability; and (3) meticulously curated, 100 million-scale datasets augmented with task-specific reward models to refine generation and editing objectives. By demonstrating that high-fidelity multimodal integration need not incur prohibitive resource demands, Skywork UniPic establishes a practical paradigm for deployable, high-fidelity multimodal AI. Code and weights are publicly available at https://huggingface.co/Skywork/Skywork-UniPic-1.5B.
title Skywork UniPic: Unified Autoregressive Modeling for Visual Understanding and Generation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2508.03320