Testing Large Language Models on Compositionality and Inference with Phrase-Level Adjective-Noun Entailment

1 code implementation • COLING 2022 • Lorenzo Bertolini, Julie Weeds, David Weir

Here, we investigate whether lexical entailment (LE, i. e. hyponymy or the is a relation between words) can be generalised in a compositional manner.

Lexical Entailment Transfer Learning

Dreams Are More "Predictable'' Than You Think

1 code implementation • 8 May 2023 • Lorenzo Bertolini

A consistent body of evidence suggests that dream reports significantly vary from other types of textual transcripts with respect to semantic content.

Automatic Scoring of Dream Reports' Emotional Content with Large Language Models

1 code implementation • 28 Feb 2023 • Lorenzo Bertolini, Valentina Elce, Adriana Michalak, Giulio Bernardi, Julie Weeds

In this work, we address these limitations by adopting large language models (LLMs) to study and replicate the manual annotation of dream reports, using a mixture of off-the-shelf and bespoke approaches, with a focus on references to reports' emotions.

text-classification Text Classification

Representing Syntax and Composition with Geometric Transformations

1 code implementation • Findings (ACL) 2021 • Lorenzo Bertolini, Julie Weeds, David Weir, Qiwei Peng

The exploitation of syntactic graphs (SyGs) as a word's context has been shown to be beneficial for distributional semantic models (DSMs), both at the level of individual word representations and in deriving phrasal representations via composition.

Knowledge Graphs

Data Augmentation for Hypernymy Detection

1 code implementation • EACL 2021 • Thomas Kober, Julie Weeds, Lorenzo Bertolini, David Weir

The automatic detection of hypernymy relationships represents a challenging problem in NLP.

Data Augmentation