UniCompress：面向统一视觉-语言理解与生成的词元压缩
UniCompress: Token Compression for Unified Vision-Language Understanding and Generation
对做统一视觉-语言模型的人，这套带可学习全局元词元的即插即用压缩/解压模块可以直接插到已有的离散词元自回归框架里，把图像词元砍到 1/4 而不必全量重训，是降低多模态推理延迟与显存的一条低成本路径。
原文依据 we propose a unified token compression algorithm UniCompress that significantly reduces visual token count while preserving performance
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论统一模型的目标是将图像编码为离散词元，并在单一自回归框架内与文本一同处理，从而同时支持理解与生成。这种统一设计带来了架构简洁性与跨模态协同效应，有利于共享参数化、一致的训练目标以及模态间的无缝迁移。然而，此类模型所需的视觉词元数量庞大，带来可观的计算与内存开销，这种低效直接阻碍了其在具身智能系统等资源受限场景中的部署。在本工作中，我们提出统一词元压缩算法 UniCompress，在保持图像理解与生成任务性能的同时显著减少视觉词元数量。该方法引入一种由可学习全局元词元引导的即插即用压缩与解压机制。该框架轻量且模块化，可在无需全量重训练的情况下高效集成到现有模型中。实验结果表明，我们的方法将图像词元最多减少 4 倍，在推理延迟与训练成本上取得显著收益，且仅带来极小的性能下降，这体现了词元高效统一建模在真实世界多模态应用中的潜力。
英文原文摘要 1346 字符
Unified models aim to support both understanding and generation by encoding images into discrete tokens and processing them alongside text within a single autoregressive framework. This unified design offers architectural simplicity and cross-modal synergy, which facilitates shared parameterization, consistent training objectives, and seamless transfer between modalities. However, the large number of visual tokens required by such models introduces substantial computation and memory overhead, and this inefficiency directly hinders deployment in resource constrained scenarios such as embodied AI systems. In this work, we propose a unified token compression algorithm UniCompress that significantly reduces visual token count while preserving performance on both image understanding and generation tasks. Our method introduces a plug-in compression and decompression mechanism guided with learnable global meta tokens. The framework is lightweight and modular, enabling efficient integration into existing models without full retraining. Experimental results show that our approach reduces image tokens by up to (4x), achieves substantial gains in inference latency and training cost, and incurs only minimal performance degradation, which demonstrates the promise of token-efficient unified modeling for real world multimodal applications.