CaLR：用于稳健扩散推理的因果潜在修正
CaLR: Causal Latent Revision for Robust Diffusion Reasoning
- Peking University（北京大学）
- Institute of Artificial Intelligence (TeleAI), China Telecom（中国电信人工智能研究院（TeleAI））
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
提出CaLR框架，结合AR与DLM优势，通过因果拓扑矩阵和隐式微分进行梯度引导思维修正，实现并行生成中逻辑一致的自校正，在复杂基准和数独任务上达到SOTA性能。
AI中文摘要：
自回归（AR）模型存在局部贪婪问题，而扩散语言模型（DLMs）通常缺乏推理所需的严格因果结构。为结合两者优势并克服各自缺点，我们提出因果潜在修正（CaLR），一个将推理重构为约束潜在优化的框架。通过采用专家模型的因果拓扑矩阵（CTM）和隐式微分，CaLR执行梯度引导的“思维修正”以强制逻辑一致性，从而在并行生成过程中实现中间步骤的动态自校正。实验上，CaLR在复杂基准上达到DLM最先进（SOTA）性能，超越强AR基线，并在数独等约束任务中展现出卓越的稳健性。
英文摘要：
Autoregressive (AR) models suffer from local greediness, while diffusion language models (DLMs) often lack the strict causal structure required for reasoning. To combine the advantages and overcome the drawbacks of the dual, we propose Causal Latent Revision (CaLR), a framework that reformulates reasoning as constrained latent optimization. By adopting a causal topology matrix (CTM) from an expert model and implicit differentiation, CaLR performs gradient-guided ``thought revision" to enforce logical consistency, enabling dynamic self-correction of intermediate steps during parallel generation. Empirically, CaLR achieves SOTA DLM performance on complex benchmarks, surpassing strong AR baselines and demonstrating superior robustness in constrained tasks like Sudoku.
