用编辑操作建模事件序列生成，提升效率与灵活性。
Edit-Based Flow Matching for Temporal Point Processes
- 通过插入、删除、替换三种编辑操作构建连续时间马尔可夫链。
- 生成时所需编辑操作总数显著减少，效率更高。
- 适合需要高效生成的事件序列建模任务，如金融、医疗数据。
时间点过程（TPPs）是建模连续时间事件序列的基础工具，但现有方法多依赖自回归参数化，受限于序列采样。近期非自回归扩散模型通过离散马尔可夫链中的事件增删实现噪声到数据的联合插值。本文提出一种新的编辑流（Edit Flow）过程，通过插入、删除和替换三种编辑操作，在连续时间马尔可夫链框架下将噪声映射至数据。通过学习瞬时编辑速率，模型具备更强灵活性与生成效率，显著降低生成过程所需的总编辑次数。实验证明，该无条件训练模型在多个基准数据集上，对无条件与条件生成任务均表现出良好的生成灵活性。
原文摘要 · Abstract (English)
Temporal point processes (TPPs) are a fundamental tool for modeling event sequences in continuous time, but most existing approaches rely on autoregressive parameterizations that are limited by their sequential sampling. Recent non-autoregressive, diffusion-style models mitigate these issues by jointly interpolating between noise and data through event insertions and deletions in a discrete Markov chain. In this work, we generalize this perspective and introduce an Edit Flow process for TPPs that transports noise to data via insert, delete, and substitute edit operations. By learning the instantaneous edit rates within a continuous-time Markov chain framework, we attain a flexible and efficient model that effectively reduces the total number of necessary edit operations during generation. Empirical results demonstrate the generative flexibility of our unconditionally trained model in a wide range of unconditional and conditional generation tasks on benchmark TPPs.
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。