A Corpus for Understanding and Generating Moral Stories

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5069 - 5087 July 10-15, 2022 ©2022 Association for Computational Linguistics A Corpus for Understanding and Generating Moral Stories Jian Guan, Ziqi Liu, Minlie Huang∗ The CoAI group, DCST; Institute for Artificial Intelligence; State Key Lab of Intelligent Technology and Systems; Beijing National Research Center for Information Science and Technology; Tsinghua University, Beijing 100084, China. {j-guan19, liuzq19}@mails.tsinghua.edu.cn, aihuang@tsinghua.edu.cn Abstract Teaching morals is one of the most important purposes of storytelling. An essential ability for understanding and writing moral stories is bridging story plots and implied morals. Its challenges mainly lie in: (1) grasping knowl- edge about abstract concepts in morals, (2) capturing inter-event discourse relations in sto- ries, and (3) aligning value preferences of sto- ries and morals concerning good or bad be- havior. In this paper, we propose two under- standing tasks and two generation tasks to as- sess these abilities of machines. We present STORAL, a new dataset of Chinese and English human-written moral stories. We show the dif- ficulty of the proposed tasks by testing vari- ous models with automatic and manual eval- uation on STORAL. Furthermore, we present a retrieval-augmented algorithm that effectively exploits related concepts or events in training sets as additional guidance to improve perfor- mance on these tasks. 1 Introduction Stories play an essential role in one’s moral devel- opment (Vitz, 1990). For example, individuals usu- ally learn morals from life experiences or literature such as fables and tell their morals by representing their lived experience in a narrative form (Tappan and Brown, 1989). Accordingly, it is a crucial ability for humans to bridge abstract morals and concrete events in stories. However, this ability has not yet been investigated for machines. There have been many tasks proposed for evalu- ating story understanding and generation, including story ending selection (Mostafazadeh et al., 2016) and story generation from short prompts (Fan et al., 2018). Unlike these tasks, which focus on reason- ing plots from context, we emphasize the ability to associate plots with implied morals. As exem- plified in Table 1, the challenges mainly lie in (1) ∗Corresponding author Stories: Four cows lived in a forest near a meadow. They were good friends and did everything together. They grazed together and stayed together, because of which no tigers or lions were able to kill them for food. But one day, the friends fought and each cow went to graze in a different direction. A tiger and a lion saw this and decided that it was the perfect opportunity to kill the cows. They hid in the bushes and surprised the cows and killed them all, one by one. Morals: Unity is strength. Table 1: An example in STORAL grasping knowledge about abstract concepts (e.g., “unity,” “strength”) and relations among them (e.g., “is”) in morals; (2) capturing inter-event discourse relations in stories (e.g., the contrast between end- ings of the “cows” when they are “united” and …

Content truncated to 500 words