If there's a Trigger Warning, then where's the Trigger? Investigating Trigger Warnings at the Passage Level
This addresses the need for more precise trigger warnings in documents for readers sensitive to harmful content, but it is incremental as it builds on existing document-level warning practices.
The paper tackled the problem of identifying specific passages in documents that prompt trigger warnings, creating a dataset of 4,135 English passages annotated with eight common warnings and evaluating fine-tuned and few-shot classifiers, finding that automatic classification is challenging but feasible.
Trigger warnings are labels that preface documents with sensitive content if this content could be perceived as harmful by certain groups of readers. Since warnings about a document intuitively need to be shown before reading it, authors usually assign trigger warnings at the document level. What parts of their writing prompted them to assign a warning, however, remains unclear. We investigate for the first time the feasibility of identifying the triggering passages of a document, both manually and computationally. We create a dataset of 4,135 English passages, each annotated with one of eight common trigger warnings. In a large-scale evaluation, we then systematically evaluate the effectiveness of fine-tuned and few-shot classifiers, and their generalizability. We find that trigger annotation belongs to the group of subjective annotation tasks in NLP, and that automatic trigger classification remains challenging but feasible.