TellMeWhy: A Dataset for Answering Why-Questions in Narratives
This addresses the problem of narrative understanding for AI researchers by providing a benchmark to evaluate and improve QA models on 'why' questions, though it is incremental as it builds on existing QA datasets.
The authors introduced TellMeWhy, a dataset of over 30k questions and answers about why characters act in narratives, with one-third requiring external commonsense knowledge, and found that state-of-the-art models perform far below humans, especially on external-answer questions.
Answering questions about why characters perform certain actions is central to understanding and reasoning about narratives. Despite recent progress in QA, it is not clear if existing models have the ability to answer "why" questions that may require commonsense knowledge external to the input narrative. In this work, we introduce TellMeWhy, a new crowd-sourced dataset that consists of more than 30k questions and free-form answers concerning why characters in short narratives perform the actions described. For a third of this dataset, the answers are not present within the narrative. Given the limitations of automated evaluation for this task, we also present a systematized human evaluation interface for this dataset. Our evaluation of state-of-the-art models show that they are far below human performance on answering such questions. They are especially worse on questions whose answers are external to the narrative, thus providing a challenge for future QA and narrative understanding research.