论文标题
很少发射文档级别的关系提取
Few-Shot Document-Level Relation Extraction
论文作者
论文摘要
我们提出弗雷多（Fredo），几张镜头级的关系提取（FSDLRE）基准。与基于句子级别的关系提取语料库建立的现有基准相反，我们认为文档级别的语料库提供了更多的现实主义，尤其是关于无原始（NOTA）分布的现实主义。因此，我们建议一组FSDLRE任务，并基于两个现有的监督学习数据集（DOCRED和SCIERC）构建基准。我们将最先进的句子级方法MNAV调整到文档级别，并进一步开发它以改进域的适应性。我们发现FSDLRE是一个充满挑战的环境，具有有趣的新特征，例如从支持集中对NOTA实例进行采样的能力。数据，代码和训练的模型可在线获得（https://github.com/nicpopovic/fredo）。
We present FREDo, a few-shot document-level relation extraction (FSDLRE) benchmark. As opposed to existing benchmarks which are built on sentence-level relation extraction corpora, we argue that document-level corpora provide more realism, particularly regarding none-of-the-above (NOTA) distributions. Therefore, we propose a set of FSDLRE tasks and construct a benchmark based on two existing supervised learning data sets, DocRED and sciERC. We adapt the state-of-the-art sentence-level method MNAV to the document-level and develop it further for improved domain adaptation. We find FSDLRE to be a challenging setting with interesting new characteristics such as the ability to sample NOTA instances from the support set. The data, code, and trained models are available online (https://github.com/nicpopovic/FREDo).
