Exploring Italian sentence embeddings properties through multi-tasking

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Nastase, Vivi, Samo, Giuseppe, Jiang, Chunyang, Merlo, Paola
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916499161088000
author Nastase, Vivi
Samo, Giuseppe
Jiang, Chunyang
Merlo, Paola
author_facet Nastase, Vivi
Samo, Giuseppe
Jiang, Chunyang
Merlo, Paola
contents We investigate to what degree existing LLMs encode abstract linguistic information in Italian in a multi-task setting. We exploit curated synthetic data on a large scale -- several Blackbird Language Matrices (BLMs) problems in Italian -- and use them to study how sentence representations built using pre-trained language models encode specific syntactic and semantic information. We use a two-level architecture to model separately a compression of the sentence embeddings into a representation that contains relevant information for a task, and a BLM task. We then investigate whether we can obtain compressed sentence representations that encode syntactic and semantic information relevant to several BLM tasks. While we expected that the sentence structure -- in terms of sequence of phrases/chunks -- and chunk properties could be shared across tasks, performance and error analysis show that the clues for the different tasks are encoded in different manners in the sentence embeddings, suggesting that abstract linguistic notions such as constituents or thematic roles does not seem to be present in the pretrained sentence embeddings.
format Preprint
id arxiv_https___arxiv_org_abs_2409_06622
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Exploring Italian sentence embeddings properties through multi-tasking
Nastase, Vivi
Samo, Giuseppe
Jiang, Chunyang
Merlo, Paola
Computation and Language
68T50
I.2.7
We investigate to what degree existing LLMs encode abstract linguistic information in Italian in a multi-task setting. We exploit curated synthetic data on a large scale -- several Blackbird Language Matrices (BLMs) problems in Italian -- and use them to study how sentence representations built using pre-trained language models encode specific syntactic and semantic information. We use a two-level architecture to model separately a compression of the sentence embeddings into a representation that contains relevant information for a task, and a BLM task. We then investigate whether we can obtain compressed sentence representations that encode syntactic and semantic information relevant to several BLM tasks. While we expected that the sentence structure -- in terms of sequence of phrases/chunks -- and chunk properties could be shared across tasks, performance and error analysis show that the clues for the different tasks are encoded in different manners in the sentence embeddings, suggesting that abstract linguistic notions such as constituents or thematic roles does not seem to be present in the pretrained sentence embeddings.
title Exploring Italian sentence embeddings properties through multi-tasking
topic Computation and Language
68T50
I.2.7
url https://arxiv.org/abs/2409.06622