DiCoRe: Enhancing Zero-shot Event Detection via Divergent-Convergent LLM Reasoning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Parekh, Tanmay, Mehta, Kartik, Mehrabi, Ninareh, Chang, Kai-Wei, Peng, Nanyun
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912591022915584
author Parekh, Tanmay
Mehta, Kartik
Mehrabi, Ninareh
Chang, Kai-Wei
Peng, Nanyun
author_facet Parekh, Tanmay
Mehta, Kartik
Mehrabi, Ninareh
Chang, Kai-Wei
Peng, Nanyun
contents Zero-shot Event Detection (ED), the task of identifying event mentions in natural language text without any training data, is critical for document understanding in specialized domains. Understanding the complex event ontology, extracting domain-specific triggers from the passage, and structuring them appropriately overloads and limits the utility of Large Language Models (LLMs) for zero-shot ED. To this end, we propose DiCoRe, a divergent-convergent reasoning framework that decouples the task of ED using Dreamer and Grounder. Dreamer encourages divergent reasoning through open-ended event discovery, which helps to boost event coverage. Conversely, Grounder introduces convergent reasoning to align the free-form predictions with the task-specific instructions using finite-state machine guided constrained decoding. Additionally, an LLM-Judge verifies the final outputs to ensure high precision. Through extensive experiments on six datasets across five domains and nine LLMs, we demonstrate how DiCoRe consistently outperforms prior zero-shot, transfer-learning, and reasoning baselines, achieving 4-7% average F1 gains over the best baseline -- establishing DiCoRe as a strong zero-shot ED framework.
format Preprint
id arxiv_https___arxiv_org_abs_2506_05128
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DiCoRe: Enhancing Zero-shot Event Detection via Divergent-Convergent LLM Reasoning
Parekh, Tanmay
Mehta, Kartik
Mehrabi, Ninareh
Chang, Kai-Wei
Peng, Nanyun
Computation and Language
Artificial Intelligence
Machine Learning
Zero-shot Event Detection (ED), the task of identifying event mentions in natural language text without any training data, is critical for document understanding in specialized domains. Understanding the complex event ontology, extracting domain-specific triggers from the passage, and structuring them appropriately overloads and limits the utility of Large Language Models (LLMs) for zero-shot ED. To this end, we propose DiCoRe, a divergent-convergent reasoning framework that decouples the task of ED using Dreamer and Grounder. Dreamer encourages divergent reasoning through open-ended event discovery, which helps to boost event coverage. Conversely, Grounder introduces convergent reasoning to align the free-form predictions with the task-specific instructions using finite-state machine guided constrained decoding. Additionally, an LLM-Judge verifies the final outputs to ensure high precision. Through extensive experiments on six datasets across five domains and nine LLMs, we demonstrate how DiCoRe consistently outperforms prior zero-shot, transfer-learning, and reasoning baselines, achieving 4-7% average F1 gains over the best baseline -- establishing DiCoRe as a strong zero-shot ED framework.
title DiCoRe: Enhancing Zero-shot Event Detection via Divergent-Convergent LLM Reasoning
topic Computation and Language
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2506.05128