PulseAugur
中
实时 22:02:40
English(EN) Lightweight Person-Place Relation Extraction from Historical Newspapers with Dependency Graphs and Proximity Features

轻量级系统从历史报纸中抽取人物地名关系

DS@GT HIPE 的研究人员开发了一个轻量级系统,用于从历史报纸中抽取人物地名关系,重点关注可解释性和效率。他们的方法利用依存图和邻近特征,使用小型 scikit-learn 集成或紧凑型图注意力网络进行关系分类,模型参数保持在 847K 以下。该系统在 HIPE-2026 共享任务上实现了 0.5142 的宏召回率,在效率方面排名第三,并强调了最小字符距离对于捕捉分类信号的重要性。研究还强调了文档分组交叉验证对于防止数据泄露的必要性。 AI

影响 这项研究展示了一种高效且可解释的关系抽取方法,有望降低历史文本分析的计算成本。

排序理由 学术论文,详细介绍了一种新的关系抽取方法。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

轻量级系统从历史报纸中抽取人物地名关系

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
学术论文,详细介绍了一种新的关系抽取方法。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
paper, other
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
103 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准。

报道来源 [1]

  1. arXiv cs.CL TIER_1 English(EN) · Mlen-Too Wesley ·

    基于依赖图和邻近特征的报刊历史文献轻量级人物地名关系抽取

    arXiv:2607.19718v1 Announce Type: new Abstract: The HIPE-2026 shared task introduces person-place relation extraction from multilingual historical newspapers as a new evaluation track, classifying the at and isAt relations between pre-annotated person and location mentions in Eng…