arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.01608cs.AI

基于自回归扩散协同思想的推理

Reasoning with Autoregressive-Diffusion Collaborative Thoughts

The Chinese University of Hong Kong（香港中文大学）；University of Science and Technology of China（中国科学技术大学）；Tsinghua University（清华大学）

Mu Yuan, Liekang Zeng, Guoliang Xing, Lan Zhang, Yunhao Liu

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出协同思想框架，通过自回归和扩散模型的闭环协作提升空间推理可靠性与生成可控性。

摘要：

自回归模型和扩散模型代表了两种互补的生成范式。自回归模型在顺序规划和约束组合方面表现出色，但难以处理需要显式空间或物理基础的任务。相比之下，扩散模型通过高维生成捕捉丰富的空间结构，但缺乏逐步逻辑控制来满足复杂多阶段约束或可靠地识别和纠正错误的能力。我们引入了协同思想，一种统一的协作框架，使自回归和扩散模型能够通过闭环交互共同推理和生成。在协同思想中，自回归模型执行结构化规划和约束管理，扩散模型将这些约束作为中间视觉思想进行实例化，而基于视觉的批评模块则评估这些视觉思想是否满足预期的结构和物理要求。这种反馈随后被用来迭代地优化后续的规划和生成步骤，从而在模态之间减少错误传播。重要的是，协同思想无论任务是自回归问答还是基于扩散的视觉生成，都使用相同的协作循环。通过代表性示例，我们展示了协同思想如何提高空间推理的可靠性以及生成的可控性。

英文摘要：

Autoregressive and diffusion models represent two complementary generative paradigms. Autoregressive models excel at sequential planning and constraint composition, yet struggle with tasks that require explicit spatial or physical grounding. Diffusion models, in contrast, capture rich spatial structure through high-dimensional generation, but lack the stepwise logical control needed to satisfy complex, multi-stage constraints or to reliably identify and correct errors. We introduce Collaborative Thoughts, a unified collaborative framework that enables autoregressive and diffusion models to reason and generate jointly through a closed-loop interaction. In Collaborative Thoughts, autoregressive models perform structured planning and constraint management, diffusion models instantiate these constraints as intermediate visual thoughts, and a vision-based critic module evaluates whether the visual thoughts satisfy the intended structural and physical requirements. This feedback is then used to iteratively refine subsequent planning and generation steps, mitigating error propagation across modalities. Importantly, Collaborative Thoughts uses the same collaborative loop regardless of whether the task is autoregressive question answering or diffusion-based visual generation. Through representative examples, we illustrate how Collaborative Thoughts can improve the reliability of spatial reasoning and the controllability of generation.

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。

↑