cff-version: 1.2.0 abstract: "

Code for the paper "What does a Text Classifier Learn about Morality? An Explainable Method for Cross-Domain Comparison of Moral Rhetoric", published at ACL '23. This code implements Tomea, an Explainable AI method for investigating the difference in how language models represent morality across domains. Given a pair of datasets and models trained on the datasets, Tomea generates 10 m-distances and one d-distance to measure the difference between the datasets, based on the SHAP method. We make pairwise comparisons of seven models trained on the MFTC datasets (available at this DOI: 10.4121/646b20e3-e24f-452d-938a-bcb6ce30913c).

" authors: - family-names: Liscio given-names: Enrico orcid: "https://orcid.org/0000-0002-8285-5867" - family-names: Araque given-names: Oscar orcid: "https://orcid.org/0000-0003-3224-0001" - family-names: Gatti given-names: Lorenzo orcid: "https://orcid.org/0000-0003-2422-5055" - family-names: Constantinescu given-names: Ionut orcid: "https://orcid.org/0009-0003-5494-0161" - family-names: Jonker given-names: C.M. (Catholijn) orcid: "https://orcid.org/0000-0003-4780-7461" - family-names: Kalimeri given-names: Kyriaki orcid: "https://orcid.org/0000-0001-8068-5916" - family-names: Murukannaiah given-names: Pradeep K. orcid: "https://orcid.org/0000-0002-1261-6908" title: "What does a Text Classifier Learn about Morality? An Explainable Method for Cross-Domain Comparison of Moral Rhetoric - code" keywords: version: 1 identifiers: - type: doi value: 10.4121/1e71138c-be26-4652-971a-48a84837df8e.v1 license: MIT date-released: 2023-12-18