Paper Detail
Jinbo Xing, Zeyinzi Jiang, Yuxiang Tuo, Chaojie Mao, Xiaotang Gai, Xi Chen, Jingfeng Zhang, Yulin Pan, Zhen Han, Jie Xiao, Keyu Yan, Chenwei Xie, Chongyang Zhong, Kai Zhu, Tong Shen, Lianghua Huang, Yu Liu, Yujiu Yang
Recent unified models have made unprecedented progress in both understanding and generation. However, while most of them accept multi-modal inputs, they typically produce only single-modality outputs. This challenge of producing interleaved content is mainly due to training data scarcity and the difficulty of modeling long-range cross-modal context. To address this issue, we decompose interleaved generation into textual planning and visual consistency modeling, and introduce a framework consisting of a planner and a visualizer. The planner produces dense textual descriptions for visual content, while the visualizer synthesizes images accordingly. Under this guidance, we construct large-scale textual-proxy interleaved data (where visual content is represented in text) to train the planner, and curate reference-guided image data to train the visualizer. These designs give rise to Wan-Weaver, which exhibits emergent interleaved generation ability with long-range textual coherence and visual consistency. Meanwhile, the integration of diverse understanding and generation data into planner training enables Wan-Weaver to achieve robust task reasoning and generation proficiency. To assess the model's capability in interleaved generation, we further construct a benchmark that spans a wide range of use cases across multiple dimensions. Extensive experiments demonstrate that, even without access to any real interleaved data, Wan-Weaver achieves superior performance over existing methods.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@article{xing2026wan,
title = {Wan-Weaver: Interleaved Multi-modal Generation via Decoupled Training},
author = {Jinbo Xing and Zeyinzi Jiang and Yuxiang Tuo and Chaojie Mao and Xiaotang Gai and Xi Chen and Jingfeng Zhang and Yulin Pan and Zhen Han and Jie Xiao and Keyu Yan and Chenwei Xie and Chongyang Zhong and Kai Zhu and Tong Shen and Lianghua Huang and Yu Liu and Yujiu Yang},
year = {2026},
abstract = {Recent unified models have made unprecedented progress in both understanding and generation. However, while most of them accept multi-modal inputs, they typically produce only single-modality outputs. This challenge of producing interleaved content is mainly due to training data scarcity and the difficulty of modeling long-range cross-modal context. To address this issue, we decompose interleaved generation into textual planning and visual consistency modeling, and introduce a framework consisti},
url = {https://arxiv.org/abs/2603.25706},
keywords = {cs.CV},
eprint = {2603.25706},
archiveprefix = {arXiv},
}
{}