发表机构
German Research Center for Artificial Intelligence (DFKI); Carl von Ossietzky Universität Oldenburg（德国人工智能研究中心（DFKI）; 卡尔·冯·奥西茨基奥尔登堡大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对大型语言模型端到端事件抽取的挑战，提出EAGER框架，结合细粒度可验证奖励与模式对比优势估计，在七个基准数据集上显著优于现有方法，提升结构化抽取性能。
AI 中文摘要
端到端事件抽取对大型语言模型而言仍具挑战性，因为它需要同时识别事件触发词、分类事件类型，并抽取基于模式（schema）的参数片段。我们提出EAGER，一个用于生成式事件抽取的强化学习框架，该框架将细粒度的可验证奖励与模式对比优势估计（Schema-Contrastive Advantage Estimation）相结合，以缓解稀疏二元奖励下的优势坍缩问题。我们的奖励设计明确针对结构有效性、抽取准确性、基于事实性、覆盖度、过度生成和片段精确度。在七个基准数据集上的实验表明，EAGER始终优于提示方法、监督微调以及先前的强化学习基线，相较于最强先前方法取得了显著提升。结果表明，任务对齐的可验证奖励和对比优势估计大幅提升了结构化抽取性能。
英文摘要
End-to-end event extraction remains challenging for large language models as it requires simultaneous identification of event triggers, classification of event types, and extraction of schema-grounded argument spans. We present EAGER, a reinforcement learning framework for generative event extraction that combines fine-grained verifiable rewards with Schema-Contrastive Advantage Estimation to alleviate advantage collapse under sparse binary rewards. Our reward design explicitly targets structural validity, extraction accuracy, groundedness, coverage, over-generation, and span precision. Experiments across seven benchmark datasets show that EAGER consistently outperforms prompting, supervised fine-tuning, and prior reinforcement learning baselines, achieving a substantial improvement over the strongest prior method. Results demonstrate that task-aligned verifiable rewards and contrastive advantage estimation substantially improve structured extraction.
CommentsAccepted to EMNLP 2026 Findings
