用扩散模型思路重构思维链，让大模型推理更抗错、可修正。
DiffCoT: Diffusion-styled Chain-of-Thought Reasoning in LLMs
- 将思维链生成看作逐步去噪过程，支持中间步骤的回溯修正。
- 在三个数学推理基准上均超越传统方法，错误传播显著减少。
- 适合需要高可靠性推理的场景，如数学竞赛或科学计算。
思维链（CoT）推理虽提升大模型多步数学求解能力，但仍受暴露偏差与错误累积影响，早期错误在自回归解码中不可逆传播。本文提出 DiffCoT，一种基于扩散模型思想的思维链框架，将推理过程重新建模为迭代去噪过程。通过滑动窗口机制在推理步骤层面融合扩散原理，实现生成与回溯修正的统一，同时保留词元级自回归性。为保障因果一致性，引入符合推理时序结构的因果扩散噪声调度。在三种涵盖不同模型架构的多步思维链基准上的实验表明，DiffCoT 持续优于现有 CoT 优化方法，显著增强推理鲁棒性与纠错能力。
原文摘要 · Abstract (English)
Chain-of-Thought (CoT) reasoning improves multi-step mathematical problem solving in large language models but remains vulnerable to exposure bias and error accumulation, as early mistakes propagate irreversibly through autoregressive decoding. In this work, we propose DiffCoT, a diffusion-styled CoT framework that reformulates CoT reasoning as an iterative denoising process. DiffCoT integrates diffusion principles at the reasoning-step level via a sliding-window mechanism, enabling unified generation and retrospective correction of intermediate steps while preserving token-level autoregression. To maintain causal consistency, we further introduce a causal diffusion noise schedule that respects the temporal structure of reasoning chains. Extensive experiments on three multi-step CoT reasoning benchmarks across diverse model backbones demonstrate that DiffCoT consistently outperforms existing CoT preference optimization methods, yielding improved robustness and error-correction capability in CoT reasoning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。