A Diffusion Model over Directed Acyclic Graphs for Event Schema Generation
Guoxuan Ding, Haotian Jin, Xiaobo Guo, Xin Wang, Nan Mu, Lei Wang, Daren Zha
Abstract
Event schema generation is crucial for understanding the structure and temporal relationships of complex events. In this paper, we introduce a novel Directed Acyclic Graph Diffusion Model (DAGDM) that integrates DAG characteristics within a diffusion framework to enhance the effectiveness of schema generation. Our method leverages DAG positional embeddings to capture the hierarchical structure of nodes within graphs, while employing a reachability-based attention to better extract structural relationships between events. To this end, we design a cross-generation strategy that separately generates event sequence and adjacency matrix. Experiments show that our model effectively captures long-range event sequences, significantly enhancing schema generation for complex events. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>
BibTeX
@inproceedings{icassp2025_adiffusionmodelo,
title = {A Diffusion Model over Directed Acyclic Graphs for Event Schema Generation},
author = {Guoxuan Ding and Haotian Jin and Xiaobo Guo and Xin Wang and Nan Mu and Lei Wang and Daren Zha},
booktitle = {ICASSP 2025},
year = {2025}
}