TACIT:面向隐式思维的变换感知捕捉
TACIT: Transformation-Aware Capturing of Implicit Thought
AI总结:
TACIT提出一种基于扩散的Transformer模型,通过像素空间中的校正流实现可解释的视觉推理,在迷宫求解中展示了隐式思维的同步涌现特性。
摘要:
我们提出了TACIT(面向隐式思维的变换感知捕捉),一种基于扩散的Transformer模型,用于可解释的视觉推理。与基于语言的推理系统不同,TACIT完全在像素空间中运行,利用校正流,使推理过程可以在每个推理步骤中直接可视化。我们在迷宫求解中展示了该方法,其中模型学习将未解决的迷宫图像转换为解决方案。在100万对合成迷宫数据上的关键结果包括:100个周期内训练损失减少了192倍,L2距离到真实值提高了22.7倍,仅需10个欧拉步骤(而典型扩散模型需要100-1000个)。定量分析揭示出一种显著的相变现象:在68%的变换过程中解决方案仍然不可见(零召回率),然后在t=0.70时突然出现,仅在过程的2%内。最值得注意的是,100%的样本在所有空间区域同时出现,排除了顺序路径构建的可能性,提供了整体而非算法推理的证据。这种“顿悟时刻”模式——长期酝酿后突然结晶——与人类认知中的顿悟现象相呼应。像素空间设计与无噪声流匹配为理解神经网络如何发展出在语言之前和之下运作的隐式推理策略提供了基础。
英文摘要:
We present TACIT (Transformation-Aware Capturing of Implicit Thought), a diffusion-based transformer for interpretable visual reasoning. Unlike language-based reasoning systems, TACIT operates entirely in pixel space using rectified flow, enabling direct visualization of the reasoning process at each inference step. We demonstrate the approach on maze-solving, where the model learns to transform images of unsolved mazes into solutions. Key results on 1 million synthetic maze pairs include: - 192x reduction in training loss over 100 epochs - 22.7x improvement in L2 distance to ground truth - Only 10 Euler steps required (vs. 100-1000 for typical diffusion models) Quantitative analysis reveals a striking phase transition phenomenon: the solution remains invisible for 68% of the transformation (zero recall), then emerges abruptly at t=0.70 within just 2% of the process. Most remarkably, 100% of samples exhibit simultaneous emergence across all spatial regions, ruling out sequential path construction and providing evidence for holistic rather than algorithmic reasoning. This "eureka moment" pattern -- long incubation followed by sudden crystallization -- parallels insight phenomena in human cognition. The pixel-space design with noise-free flow matching provides a foundation for understanding how neural networks develop implicit reasoning strategies that operate below and before language.
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。