What question did this study set out to answer?

The research aims to assess how well systems can understand narratives by selecting the correct ending of a story.

January 1, 2017Open Access

LSDSem 2017 Shared Task: The Story Cloze Test

View Full Paper

Ask AI

Bookmark

Authors

Shared task evaluates story understanding in systems through narrative endings implying improved performance.

The research aims to assess how well systems can understand narratives by selecting the correct ending of a story.
Evaluated eight systems in the Story Cloze Test shared task.
Systems utilized various approaches, including end-to-end neural networks, feature-based regression models, and rule-based methods.
The task involved providing each system with a four-sentence story and two possible endings to choose from.
The highest performing system achieved 75.2% accuracy.
The results represent a significant improvement over the previous state-of-the-art in narrative understanding.

Mostafazadeh et al. (2017) studied this question.

View Full Paper

Ask AI

Bookmark