Uni-cot: 向跨文本和视觉的统一链式推理迈进
Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Vision
AI总结：
Uni-CoT通过统一的链式推理框架实现跨文本和视觉的连贯多模态推理，采用宏级和微级推理范式，提升多模态推理的效率和性能。
摘要：
链式推理（CoT）已被广泛采用以通过将复杂任务分解为更简单的、顺序的子任务来增强大型语言模型（LLMs）。然而，将CoT扩展到视觉-语言推理任务仍然具有挑战性，因为这通常需要解释视觉状态的转换以支持推理。现有方法往往在这方面遇到困难，因为它们在建模视觉状态转换的能力有限，或者由于碎片化的架构导致不一致的视觉轨迹。为了克服这些限制，我们提出了Uni-CoT，一种统一的链式推理框架，使单个统一模型能够进行连贯且基于现实的多模态推理。关键思想是利用能够同时进行图像理解和生成的模型来推理视觉内容并建模演进的视觉状态。然而，使统一模型实现这一点是不平凡的，因为计算成本高且训练负担重。为此，Uni-CoT引入了一种新的两级推理范式：一个宏级CoT用于高层任务规划，一个微级CoT用于子任务执行。这种设计显著降低了计算开销。此外，我们引入了一种结构化的训练范式，将交错的图像-文本监督用于宏级CoT，与多任务目标结合用于微级CoT。这些创新使Uni-CoT能够执行可扩展且连贯的多模态推理。此外，得益于我们的设计，所有实验都可以仅使用8块80GB VRAM的A100显存的GPU高效完成。在推理驱动的图像生成基准（WISE）和编辑基准（RISE和KRIS）上的实验结果表明，Uni-CoT展示了SOTA性能和强大的泛化能力，确立了Uni-CoT作为多模态推理的有前景的解决方案。项目页面和代码：https://sais-fuxi.github.io/projects/uni-cot/
英文摘要：
Chain-of-Thought (CoT) reasoning has been widely adopted to enhance Large Language Models (LLMs) by decomposing complex tasks into simpler, sequential subtasks. However, extending CoT to vision-language reasoning tasks remains challenging, as it often requires interpreting transitions of visual states to support reasoning. Existing methods often struggle with this due to limited capacity of modeling visual state transitions or incoherent visual trajectories caused by fragmented architectures. To overcome these limitations, we propose Uni-CoT, a Unified Chain-of-Thought framework that enables coherent and grounded multimodal reasoning within a single unified model. The key idea is to leverage a model capable of both image understanding and generation to reason over visual content and model evolving visual states. However, empowering a unified model to achieve that is non-trivial, given the high computational cost and the burden of training. To address this, Uni-CoT introduces a novel two-level reasoning paradigm: A Macro-Level CoT for high-level task planning and A Micro-Level CoT for subtask execution. This design significantly reduces the computational overhead. Furthermore, we introduce a structured training paradigm that combines interleaved image-text supervision for macro-level CoT with multi-task objectives for micro-level CoT. Together, these innovations allow Uni-CoT to perform scalable and coherent multi-modal reasoning. Furthermore, thanks to our design, all experiments can be efficiently completed using only 8 A100 GPUs with 80GB VRAM each. Experimental results on reasoning-driven image generation benchmark (WISE) and editing benchmarks (RISE and KRIS) indicates that Uni-CoT demonstrates SOTA performance and strong generalization, establishing Uni-CoT as a promising solution for multi-modal reasoning. Project Page and Code: https://sais-fuxi.github.io/projects/uni-cot/
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。
