andrea-de-varda Goto Github PK

affective-interlingua

Code to reproduce the paper "The Emergence of Semantic Units in Massively Multilingual Models" (LREC-COLING 2024)

build-your-search-engine

An open-ended repository for class work on core machine learning techniques. Based on https://github.com/PeARSearch/PeARS-orchard.

cross-lingual-agreement

iconicity-datadriven

Materials for deriving data-driven iconicity measurements in the auditory modality

iconicity-deep-learning

Code employed in the deep learning-based study of iconicity in language. Details can be found in the paper (de Varda & Strapparava (2022). A Cross‐Modal and Cross‐lingual Study of Iconicity in Language: Insights From Deep Learning. Cognitive Science.

mmm-universals

Materials and code relative to the MSc thesis "Language Universals in Massively Multilingual Models: an Empirical Investigation"

prediction-resource

Scripts and data relative to the cloze probability response and predictability ratings for 205 sentences (1,726 words)

semanticscape

Code to construct SemanticScape vectors, i.e., distributional semantic vectors grounded in distance patterns between objects in natural images.

surprisal-across-languages

Code to calculate surprisal values from multilingual XGLM models.