Polygl0t/LAMBADA-poly

LAMBADA Polyglot数据集是一个多语言版本的数据集，它包含了一系列叙事文本。这些文本的特点是，如果人类主体看到了整个文本，他们能够猜出最后一个词；但如果他们只看到目标词之前的最后一个句子，则无法猜出。为了在LAMBADA上取得成功，计算模型不能仅仅依赖局部上下文，而是必须能够追踪更广泛话语中的信息。该数据集用于通过单词预测任务评估计算模型对文本的理解能力。

The LAMBADA Polyglot dataset is a multilingual version consisting of a collection of narrative texts. The characteristic of these texts is that human subjects are able to guess the last word if they are exposed to the whole text, but not if they only see the last sentence preceding the target word. To succeed on LAMBADA, computational models cannot simply rely on local context, but must be able to keep track of information in the broader discourse. This dataset is used to evaluate the text understanding capabilities of computational models through a word prediction task.