# Event Ontology Expansion via LLM-Based Conceptualization

Source: [Event Ontology Expansion via LLM-Based Conceptualization](<https://www.sota2.com/research/paper/6a3a131e206542251d7c7075>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-24T05:00:04.550Z

Authors: Weicheng Ren, Zixuan Li, Long Bai, Xiaolong Jin, Jiafeng Guo, Xueqi Cheng

Published: 2026-06-19T02:28:17.000Z

## Abstract

Event ontology expansion aims to discover emerging event types from data and extend them to appropriate positions in the existing event ontology.. Existing methods typically cluster contextualized trigger representations and attach induced clusters to the ontology based on instance-level similarity. However, ontology expansion requires concept-level semantics that characterize event types, whereas contextualized trigger representations often conflate these semantics with surface contextual variation, leading to unstable clustering and unreliable hierarchy expansion. To address this issue, we propose ConceptE, a conceptualization-enhanced framework for event ontology expansion. ConceptE first derives concept-level semantics by prompting an LLM with the sentence and event trigger, producing a concise concept name and a natural-language description. It then jointly encodes these semantics with trigger information to build concept-enhanced representations aligned with ontology-level reasoning. This representation design supports more coherent event clustering, more reliable hierarchy expansion, and ontology-consistent type naming. Experiments on ACE, ERE, and MAVEN demonstrate that ConceptE consistently outperforms state-of-the-art approaches across all subtasks of event ontology expansion. In particular, it achieves improvements of up to 12.37\\% in BCubed-F1 for event clustering and 6.48\\% in Taxo\\\_F1 for hierarchy expansion, demonstrating the effectiveness of the proposed ConceptE method.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Hierarchy Expansion on ACE](<https://www.sota2.com/research/sota/hierarchy-expansion-on-ace.md>): Hierarchy expansion; ACE; Taxonomy Precision: 55.

- [Hierarchy expansion on ERE](<https://www.sota2.com/research/sota/hierarchy-expansion-on-ere.md>): Hierarchy expansion; ERE; Taxo Precision: 32.

- [Hierarchy Expansion on MAVEN](<https://www.sota2.com/research/sota/hierarchy-expansion-on-maven.md>): Hierarchy expansion; MAVEN; Taxonomy Precision: 43.75.

- [Event Type Naming on ACE](<https://www.sota2.com/research/sota/event-type-naming-on-ace.md>): Event type naming; ACE; ROUGE-L: 30.43.

- [Event type naming on ERE](<https://www.sota2.com/research/sota/event-type-naming-on-ere.md>): Event type naming; ERE; ROUGE-L: 17.66.

- [Event type naming on MAVEN](<https://www.sota2.com/research/sota/event-type-naming-on-maven.md>): Event type naming; MAVEN; ROUGE-L: 33.36.

- [Event Clustering on ACE](<https://www.sota2.com/research/sota/event-clustering-on-ace.md>): Event Clustering; ACE; ARI: 76.27.

- [Event Clustering on ERE](<https://www.sota2.com/research/sota/event-clustering-on-ere.md>): Event Clustering; ERE; ARI: 65.61.

- [Event Clustering on MAVEN](<https://www.sota2.com/research/sota/event-clustering-on-maven.md>): Event Clustering; MAVEN; ARI: 46.99.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.21048>)

[PDF](<https://arxiv.org/pdf/2606.21048.pdf>)
