将思维链转为图像，让大模型推理过程更高效可追踪
Render-of-Thought: Rendering Textual Chain-of-Thought as Images for Visual Latent Reasoning
- 用视觉模型将文本思维链转为图像，实现推理过程可视化
- 实现3-4倍的令牌压缩，推理速度显著提升
- 无需额外训练，适合关注推理效率与可解释性的研究者
思维链（CoT）提示已在激发大语言模型（LLM）推理能力方面取得显著成果。尽管CoT提升了推理效果，但其冗长性带来了巨大的计算开销。现有工作多聚焦于结果对齐，缺乏对中间推理过程的监督，导致潜在推理链难以分析。为此，我们提出首个将推理链具象化的框架——渲染思维链（Render-of-Thought, RoT），通过将文本推理步骤转化为图像，使隐式推理过程显式且可追溯。具体而言，我们利用现有视觉语言模型（VLM）的视觉编码器作为语义锚点，对齐视觉嵌入与文本空间，实现即插即用，无需额外预训练开销。在数学与逻辑推理基准上的大量实验表明，该方法相比显式CoT实现3-4倍令牌压缩，并带来显著推理加速，同时保持与主流方法相当的性能，验证了该范式的可行性。代码已开源：https://github.com/TencentBAC/RoT
原文摘要 · Abstract (English)
Chain-of-Thought (CoT) prompting has achieved remarkable success in unlocking the reasoning capabilities of Large Language Models (LLMs). Although CoT prompting enhances reasoning, its verbosity imposes substantial computational overhead. Recent works often focus exclusively on outcome alignment and lack supervision on the intermediate reasoning process. These deficiencies obscure the analyzability of the latent reasoning chain. To address these challenges, we introduce Render-of-Thought (RoT), the first framework to reify the reasoning chain by rendering textual steps into images, making the latent rationale explicit and traceable. Specifically, we leverage the vision encoders of existing Vision Language Models (VLMs) as semantic anchors to align the vision embeddings with the textual space. This design ensures plug-and-play implementation without incurring additional pre-training overhead. Extensive experiments on mathematical and logical reasoning benchmarks demonstrate that our method achieves 3-4x token compression and substantial inference acceleration compared to explicit CoT. Furthermore, it maintains competitive performance against other methods, validating the feasibility of this paradigm. Our code is available at https://github.com/TencentBAC/RoT
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。