Learning to Generate and Extract: A Multi-Agent Collaboration Framework For Zero-shot Document-level Event Arguments Extraction

2026-03-03

AI Summary

This work addresses the challenge of zero-shot document-level event argument extraction, where scarce annotations and the inability of existing synthetic data to accurately capture contextual and structural relationships of unseen events hinder performance. To overcome these limitations, the authors propose a multi-agent collaborative framework that introduces, for the first time, a “propose–evaluate–revise” cognitive pipeline: a generator agent synthesizes data, while an evaluator agent assesses it against event-structural constraints and semantic consistency. Both agents are jointly optimized via reinforcement learning. The approach significantly improves synthetic data quality and downstream extraction performance, achieving state-of-the-art results across three zero-shot settings on RAMS and WikiEvents. Moreover, the generated data effectively enhances the zero-shot capabilities of other document-level event argument extraction models.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Multiagent Systems: Adversarial Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI

Abstract

Document-level event argument extraction (DEAE) is essential for knowledge acquisition, aiming to extract participants of events from documents.In the zero-shot setting, existing methods employ LLMs to generate synthetic data to address the challenge posed by the scarcity of annotated data. However, relying solely on Event-type-only prompts makes it difficult for the generated content to accurately capture the contextual and structural relationships of unseen events. Moreover, ensuring the reliability and usability of synthetic data remains a significant challenge due to the absence of quality evaluation mechanisms. To this end, we introduce a multi-agent collaboration framework for zero-shot document-level event argument extraction (ZS-DEAE), which simulates the human collaborative cognitive process of "Propose-Evaluate-Revise." Specifically, the framework comprises a generation agent and an evaluation agent. The generation agent synthesizes data for unseen events by leveraging knowledge from seen events, while the evaluation agent extracts arguments from the synthetic data and assesses their semantic consistency with the context. The evaluation results are subsequently converted into reward signals, with event structure constraints incorporated into the reward design to enable iterative optimization of both agents via reinforcement learning.In three zero-shot scenarios constructed from the RAMS and WikiEvents datasets, our method achieves improvements both in data generation quality and argument extraction performance, while the generated data also effectively enhances the zero-shot performance of other DEAE models.

Problem

• zero-shot
• document-level event argument extraction
• synthetic data
• data quality
• event structure

Innovation

• multi-agent collaboration
• zero-shot event argument extraction
• synthetic data generation
• reinforcement learning
• semantic consistency evaluation