arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2408.04873cs.CL
收藏

将无监督情节检测与 LLM 协同用于大规模新闻事件

Synergizing Unsupervised Episode Detection with LLMs for Large-Scale News Events

Siebel School of Computing and Data Science(塞比尔计算与数据科学学院);University of Illinois at Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Priyanka Kargupta, Yunyi Zhang, Yizhu Jiao, Siru Ouyang, Jiawei Han

AI总结:

提出无监督情节检测任务与 EpiMine 框架,利用判别词组合变化划分候选情节,并结合 LLM 精炼,在三个真实数据集上平均提升 59.2%。

摘要:

最先进的自动事件检测在可解释性以及对不断演变的大规模关键事件的适应性方面存在困难——而情节结构在这些方面表现出色。情节常被忽视,它表示核心实体在特定时间和地点执行动作的内聚簇;部分有序的情节序列可以表示一个关键事件。本文提出了一项新任务,即情节检测,用于在关键事件文章的新闻语料库中识别情节。检测情节面临独特挑战,因为它们缺乏明确的时间或位置标记,而且不能仅依靠语义相似性进行合并。尽管大型语言模型(LLM)可以帮助应对这些推理难题,但它们难以处理新闻语料库中典型的长上下文。为应对这些挑战,我们提出 EpiMine,这是一个无监督框架,通过利用文章中自然的情节划分来识别关键事件的候选情节,这些划分通过判别性词语组合的变化来估计。这些候选情节更具内聚性,也更能代表真实情节,并与 LLM 协同,以更好地解释它们并将其精炼为最终情节。我们将 EpiMine 应用于三个在情节层面标注的多样化真实事件数据集;与基线相比,它在所有指标上实现了 59.2% 的平均提升。

英文摘要:

State-of-the-art automatic event detection struggles with interpretability and adaptability to evolving large-scale key events -- unlike episodic structures, which excel in these areas. Often overlooked, episodes represent cohesive clusters of core entities performing actions at a specific time and location; a partially ordered sequence of episodes can represent a key event. This paper introduces a novel task, episode detection, which identifies episodes within a news corpus of key event articles. Detecting episodes poses unique challenges, as they lack explicit temporal or locational markers and cannot be merged using semantic similarity alone. While large language models (LLMs) can aid with these reasoning difficulties, they suffer with long contexts typical of news corpora. To address these challenges, we introduce EpiMine, an unsupervised framework that identifies a key event's candidate episodes by leveraging natural episodic partitions in articles, estimated through shifts in discriminative term combinations. These candidate episodes are more cohesive and representative of true episodes, synergizing with LLMs to better interpret and refine them into final episodes. We apply EpiMine to our three diverse, real-world event datasets annotated at the episode level, where it achieves a 59.2% average gain across all metrics compared to baselines.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑