Ultra-fine Entity Typing with Indirect Supervision from Natural Language Inference
Ultra-fine Entity Typing with Indirect Supervision from Natural Language Inference
复制标题
DOI：
10.1162/tacl_a_00479
复制
发表时间：
2022-02
影响因子：
10.9
通讯作者：
Bangzheng Li;Wenpeng Yin;Muhao Chen
中科院分区：
文献类型：
--
作者：
Bangzheng Li;Wenpeng Yin;Muhao Chen
超精细实体分类（UFET）的任务旨在预测不同的和自由形式的单词或短语，描述句子中提到的适当类型的实体。这一任务的一个关键挑战在于大量的类型和每种类型的注释数据稀缺。现有系统将任务制定为多路分类问题，并直接或远程训练监督分类器。这导致两个问题：（i）分类器不捕获类型语义，因为类型通常被转换成索引;（ii）以这种方式开发的系统限于在预定义的类型集合内进行预测，并且通常不能推广到在训练中很少看到或看不到的类型。这项工作提出了LITE排序，一种新的方法，将实体类型化为自然语言推理（NLI）问题，利用（i）NLI的间接监督来推断有意义地表示为文本假设的类型信息并缓解数据稀缺问题，以及（ii）学习排名目标以避免预先定义类型集。实验表明，在有限的训练数据下，LITE在UFET任务上获得了最先进的性能。此外，LITE不仅在其他细粒度实体类型基准测试中获得了最佳结果，更重要的是，预训练的LITE系统在包含未知类型的新数据上运行良好。
The task of ultra-fine entity typing (UFET) seeks to predict diverse and free-form words or phrases that describe the appropriate types of entities mentioned in sentences. A key challenge for this task lies in the large number of types and the scarcity of annotated data per type. Existing systems formulate the task as a multi-way classification problem and train directly or distantly supervised classifiers. This causes two issues: (i) the classifiers do not capture the type semantics because types are often converted into indices; (ii) systems developed in this way are limited to predicting within a pre-defined type set, and often fall short of generalizing to types that are rarely seen or unseen in training. This work presents LITE🍻, a new approach that formulates entity typing as a natural language inference (NLI) problem, making use of (i) the indirect supervision from NLI to infer type information meaningfully represented as textual hypotheses and alleviate the data scarcity issue, as well as (ii) a learning-to-rank objective to avoid the pre-defining of a type set. Experiments show that, with limited training data, LITE obtains state-of-the-art performance on the UFET task. In addition, LITE demonstrates its strong generalizability by not only yielding best results on other fine-grained entity typing benchmarks, more importantly, a pre-trained LITE system works well on new data containing unseen types.1