Modeling Naive Psychology of Characters in Simple Commonsense Stories

Quick links:  [download the data]   [contact us]   [paper]

Contact one of the authors: Hannah, Antoine and/or Maarten.


Abstract

Understanding a narrative requires reading between the lines and reasoning about the unspoken but obvious implications about events and people’s mental states — a capability that is trivial for humans but remarkably hard for machines. To facilitate research addressing this challenge, we introduce a new annotation framework to explain naive psychology of story characters as fully-specified chains of mental states with respect to motivations and emotional reactions. Our work presents a new large-scale dataset with rich low-level annotations and establishes baseline performance on several new tasks, suggesting avenues for future research.

Enabling reasoning about the cause and effect of mental state changes of characters in a story.


Browsing tool

Select one of our stories and click on individual characters to see our annotations. For dev/test stories, hover over categories to see descriptions.

Theories of Motivation (Maslow and Reiss) and Emotional Reaction (Plutchik).

Story Annotation Key
char appears in line.
char doesn't appears in line.

char has a mental state change.
char doesn't have a mental state change.

Read our paper for more:
Hannah Rashkin, Antoine Bosselut, Maarten Sap, Kevin Knight & Yejin Choi (2018).
Modeling Naive Psychology of Characters in Simple Commonsense Stories. ACL [view pdf]