Latent Block-Diffusion Temporal Point Processes: A Semi-Autoregressive Framework for Asynchronous Event Sequence Generation

2026-06-23

AI Summary

This work addresses the limitations of existing approaches for multistep event sequence generation: autoregressive models suffer from error accumulation, while non-autoregressive diffusion models struggle with variable-length outputs. To overcome these challenges, the authors propose a semi-autoregressive temporal point process framework that models event blocks in a latent space and incorporates an intra-block Gaussian diffusion mechanism. This approach uniquely integrates latent-space block diffusion with block-level autoregressive structure, enabling flexible output lengths without compromising generation quality. Theoretical analysis demonstrates that block-wise generation effectively mitigates error propagation and provides Wasserstein error bounds. Extensive experiments on six real-world datasets show that the method outperforms state-of-the-art models in both unconditional and conditional generation tasks, while also revealing a trade-off between block size and generation fidelity.

Abstract

Modeling and sampling from the underlying distribution of asynchronous event sequences are crucial in various real-world applications, including social networks, medical diagnosis, and financial transactions. Existing autoregressive methods suffer from error accumulation during multi-step generation, while non-autoregressive diffusion methods are typically limited to fixed-length output sequences. In this paper, we propose Latent Block-Diffusion Temporal Point Processes (LBDTPP), a novel semi-autoregressive TPP framework that introduces a latent block diffusion mechanism for high-quality and variable-length event sequence generation. The core idea is to define an autoregressive probability distribution over event blocks in latent space and perform Gaussian diffusion within each block. By sequentially generating blocks while simultaneously sampling events in each block, LBDTPP preserves the length flexibility of autoregressive TPPs and inherits the parallel high-quality generation capability of diffusion models. Theoretically, we derive Wasserstein error bounds showing that, under suitable local approximation and prefix-stability assumptions, block-wise generation can reduce error accumulation compared with event-wise autoregressive generation. Extensive experiments on six real-world benchmark datasets demonstrate that LBDTPP outperforms state-of-the-art TPP baselines in both unconditional and conditional generation tasks. Further empirical analyses verify the benefits of latent-space diffusion and block-wise generation, and reveal the trade-off between generation quality and block size. Our code is available at https://github.com/Zh-Shuai/LBDTPP.

Problem

• asynchronous event sequences
• autoregressive generation
• diffusion models
• temporal point processes
• variable-length generation

Innovation

• semi-autoregressive
• latent block diffusion
• temporal point processes
• asynchronous event sequences
• variable-length generation