用迭代纠错机制让Transformer自适应计算深度，解数独表现接近人类水平。
Liquid Reasoning Transformers: A Sudoku-Based Prototype for Chess-Scale Algorithmic Tasks
- 通过循环更新推理标记，动态调整计算过程
- 数独任务中达98.68%数字准确率，36.30%完整解题率
- 适合需要深度推理的复杂任务，如国际象棋级问题
液态推理Transformer（LRT）是一种基于迭代更新、丢弃修正和可学习停止机制的Transformer架构，用于自适应深度推理。模型不依赖单次前向传播，而是在多个内部步骤中持续更新一个循环推理标记，从而纠正早期错误并根据输入难度分配计算量。我们在数独这一结构化推理的受控测试平台上评估LRT，结果表明其在未使用符号规则或搜索的情况下，实现了98.68%的数字准确率和36.30%的完整谜题准确率。内部分析显示，丢弃门和停止门在稳定推断与调节计算深度方面发挥不同且关键的作用。我们讨论这些机制如何自然扩展至国际象棋级别的推理任务，并提出了多标记推理及更大领域应用的拓展方向。
原文摘要 · Abstract (English)
The Liquid Reasoning Transformer (LRT) is a transformer architecture designed for inference with adaptive depths using iterative changes, discard-based correction, and a learned stopping mechanism. Instead of relying on a single feedforward pass, the model updates a recurrent reasoning token across multiple internal steps, allowing it to correct early errors and allocate computation based on input difficulty. We evaluate the LRT on Sudoku as a controlled testbed for structured reasoning and show that it achieves strong performance, reaching 98.68% digit accuracy and 36.30% full-puzzle accuracy without using symbolic rules or search. Analyzing internal patterns shows that the discard and stop gates play different, important roles in stabilizing inferences and adjusting computational depth. We discuss how these mechanisms extend naturally to chess-scale reasoning tasks and outline extensions for multi-token reasoning and larger domains.
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。