---
title: "A Scalable Cross-Domain Event Extraction System via a Unified Generative Training Framework"
authors:
  - "Siting Liang"
  - "Omar Adjali"
  - "Omair Shahzad Bhatti"
  - "Daniel Sonntag"
doi: "10.24963/ijcai.2026/977"
url: "https://ijcai.org/proceedings/2026/977"
pdf: "https://ijcai.org/proceedings/2026/0977.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "8449-8452"
track: "Demo Track"
keywords:
  - "AI: Humans and AI"
  - "AI: Machine Learning"
  - "AI: Natural Language Processing"
---

# A Scalable Cross-Domain Event Extraction System via a Unified Generative Training Framework

Siting Liang, Omar Adjali, Omair Shahzad Bhatti, Daniel Sonntag

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 8449-8452. DOI: [10.24963/ijcai.2026/977](https://doi.org/10.24963/ijcai.2026/977)

## Abstract

Event extraction is fundamental to information extraction. Prior approaches often separate event detection and argument extraction or depend on dataset-specific designs, limiting scalability and cross-domain generalization. We propose a unified generative, sequence-to-sequence framework that performs all event extraction subtasks jointly and supports both end-to-end and pipeline configurations. We fine-tune pre-trained language models on multiple event datasets across diverse domains, enabling a single model to retain domain-specific semantics while generalizing over large, evolving label spaces. Cross-domain experiments show strong, robust performance across datasets, demonstrating a scalable solution for real-world event extraction. We demonstrate these capabilities through a web-based application tailored for researchers and practitioners. The platform supports inspection of different configurations and facilitates cross-domain comparisons.