Lance：通过多任务协同实现统一多模态建模
Lance: Unified Multimodal Modeling by Multi-Task Synergy
AI总结：
本文提出Lance，一种轻量级的原生统一模型，支持图像和视频的多模态理解、生成和编辑。该模型通过协同多任务训练的实用范式实现统一多模态建模，基于统一上下文建模和解耦能力路径两个核心原则，通过双流混合专家架构实现联合上下文学习并解耦理解和生成路径。
摘要：
我们提出了Lance，一种轻量级的原生统一模型，支持图像和视频的多模态理解、生成和编辑。与依赖模型容量扩展或文本-图像主导设计不同，Lance通过协同多任务训练探索统一多模态建模的实用范式。其基于两个核心原则：统一上下文建模和解耦能力路径。具体而言，Lance从头开始训练，并在共享交错的多模态序列上采用双流混合专家架构，实现联合上下文学习的同时解耦理解和生成路径。我们进一步引入模态感知的旋转位置编码以减轻异构视觉标记之间的干扰并提升跨任务对齐。在训练过程中，Lance采用分阶段的多任务训练范式，结合能力导向的目标和自适应数据调度，以加强语义理解和视觉生成性能。实验结果表明，Lance在图像和视频生成方面显著优于现有开源统一模型，同时保留了强大的多模态理解能力。该模型的主页可在https://lance-project.github.io上访问。
英文摘要：
We present Lance, a lightweight native unified model supporting multimodal understanding, generation, and editing for both images and videos. Rather than relying on model capacity scaling or text-image-dominant designs, Lance explores a practical paradigm for unified multimodal modeling via collaborative multi-task training. It is grounded in two core principles: unified context modeling and decoupled capability pathways. Specifically, Lance is trained from scratch and employs a dual-stream mixture-of-experts architecture on shared interleaved multimodal sequences, enabling joint context learning while decoupling the pathways for understanding and generation. We further introduce modality-aware rotary positional encoding to mitigate interference among heterogeneous visual tokens and boost cross-task alignment. During training, Lance adopts a staged multi-task training paradigm with capability-oriented objectives and adaptive data scheduling to strengthen both semantic comprehension and visual generation performance. Experimental results demonstrate that Lance substantially outperforms existing open-source unified models in image and video generation, while retaining strong multimodal understanding capabilities. The homepage is available at https://lance-project.github.io.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
