Dream reports are recollections of our experiences while asleep, and have strong research and clinical value. Since their analysis can be extremely time-consuming, researchers have adopted multiple types of automatised approaches, including, in more recent years, pre-trained language models (PLMs). However, most work has focused on limited aspects of the report content, such as characters or emotions. In this work, we introduce PreDA (prefix-based dream reports annotation), a framework to build language models able to annotate a dream report for multiple relevant aspects, using sequence-to-sequence PLMs. We provide experimental evidence showing how a single PLM of small dimension can efficiently annotate a report on multiple features of the Hall and Van De Castle (HVDC) framework, give a detailed analysis of the model's performance, and explain how the training data impact learning and generalisation ability of the model.
BERTOLINI Lorenzo;
COMTE Valentin;
CERESA Mario;
CONSOLI Sergio;
2026-05-27
SPRINGER SCIENCE AND BUSINESS MEDIA DEUTSCHLAND GMBH
JRC139403
1611-3349 (online),
https://link.springer.com/chapter/10.1007/978-3-032-21477-5_13,
https://publications.jrc.ec.europa.eu/repository/handle/JRC139403,
10.1007/978-3-032-21477-5_13 (online),
This document is only visible at the Commission level.
You are not authorized to publish or distribute it outside the European Commission.
This is a public document. You can share this publication.
Additional supporting files
| File name | Description | File type | |
|---|
