Morality is Non-Binary: Building a Pluralist Moral Sentence Embedding Space using Contrastive Learning - code

DOI:10.4121/a3e5abdf-f777-4105-be40-fb9b2e3fed92.v1
The DOI displayed above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
DOI: 10.4121/a3e5abdf-f777-4105-be40-fb9b2e3fed92
Datacite citation style:
Park, Jeongwoo; Liscio, Enrico; Murukannaiah, Pradeep K. (2024): Morality is Non-Binary: Building a Pluralist Moral Sentence Embedding Space using Contrastive Learning - code. Version 1. 4TU.ResearchData. software. https://doi.org/10.4121/a3e5abdf-f777-4105-be40-fb9b2e3fed92.v1
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite

Software

We train embedding spaces with the MFTC corpus, to see how an embedding space can learn the distribution of pluralist morality. We compare off-the-shelf, unsupervised, and supervised approaches, showing that a supervised approach is necessary. Here, you can find the code used to train and evaluate the resulting embedding spaces.

History

  • 2024-01-30 first online, published, posted

Publisher

4TU.ResearchData

Format

Python code

Funding

  • Hybrid Intelligence Center (a 10-year programme funded by the Dutch Ministry of Education, Culture and Science through the Netherlands Organisation for Scientific Research).

Organizations

TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science, Department of Intelligent Systems

DATA

To access the source code, use the following command:

git clone https://data.4tu.nl/v3/datasets/296d19ac-d818-472d-a55d-568695e83d2b.git

Or download the latest commit as a ZIP.