April 26, 2024, 4:47 a.m. | Md Nayem Uddin, Enfa Rose George, Eduardo Blanco, Steven Corman

cs.CL updates on arXiv.org arxiv.org

arXiv:2404.16413v1 Announce Type: new
Abstract: This paper presents a question-answering approach to extract document-level event-argument structures. We automatically ask and answer questions for each argument type an event may have. Questions are generated using manually defined templates and generative transformers. Template-based questions are generated using predefined role-specific wh-words and event triggers from the context document. Transformer-based questions are generated using large language models trained to formulate questions based on a passage and the expected answer. Additionally, we develop novel data …

abstract arxiv cs.cl document event extract generated generative paper question questions role template transformers type words

Lead Developer (AI)

@ Cere Network | San Francisco, US

Research Engineer

@ Allora Labs | Remote

Ecosystem Manager

@ Allora Labs | Remote

Founding AI Engineer, Agents

@ Occam AI | New York

AI Engineer Intern, Agents

@ Occam AI | US

AI Research Scientist

@ Vara | Berlin, Germany and Remote