Multi agent method improves event linking accuracy in text
Enhancing Event Candidate Acquisition for Event Linking
Artificial Intelligence
Summary
Linking events mentioned in text to a knowledge base is hard because event descriptions can be short and unclear. The authors developed MACE, a method that uses multiple specialized language agents to gather clues about when, where, and who is involved in an event before trying to link it. This approach helps find better event matches without changing the linking systems themselves. Tests show that MACE improves accuracy across different event linking models.
What this means in practice
- •For natural language processing developers: Improve event extraction tools by integrating multi-agent evidence gathering for more accurate event linking.
- •For intelligence analysis teams: Enhance automated systems that link text descriptions of events to known entities, reducing errors from ambiguous mentions.
Authors
Ziyang Zhang, Yinan Liu, Boyi Xue, Yingxuan Huang, Bin Wang, Xiaochun Yang
Abstract
Event linking associates event mentions in text with entries in a knowledge base (KB), or identifies them as out-of-KB events. Although existing methods use different architectures, candidate event acquisition can still be weakened by short ambiguous mentions, noisy arguments, and evidence that is unevenly useful for retrieval. We present MACE, a Multi-Agent Candidate Event acquisition method that refines event structure before linking. MACE uses evidence-specialized LLM agents to acquire time, location, participant, and event-type evidence, exposes intermediate queries to candidate-event lookup tools, and lets a coordinator revise the evidence set before final candidate construction. Experiments on two event linking benchmarks show that adding MACE to different event linking models consistently improves accuracy. These results show that MACE improves event linking through better candidate event acquisition without modifying the event linking model.