1 code implementation • EMNLP (BlackboxNLP) 2021 • Marianna Apidianaki, Aina Garí Soler
Large scale language models encode rich commonsense knowledge acquired through exposure to massive data during pre-training, but their understanding of entities and their semantic properties is unclear.
1 code implementation • LREC 2022 • Aina Garí Soler, Matthieu Labeau, Chloé Clavel
Our discourses are full of potential lexical ambiguities, due in part to the pervasive use of words having multiple senses.
1 code implementation • COLING 2022 • Aina Garí Soler, Matthieu Labeau, Chloé Clavel
The way we use words is influenced by our opinion.
1 code implementation • 22 Feb 2024 • Aina Garí Soler, Matthieu Labeau, Chloé Clavel
When deriving contextualized word representations from language models, a decision needs to be made on how to obtain one for out-of-vocabulary (OOV) words that are segmented into subwords.
no code implementations • 12 Oct 2021 • Marianna Apidianaki, Aina Garí Soler
Large scale language models encode rich commonsense knowledge acquired through exposure to massive data during pre-training, but their understanding of entities and their semantic properties is unclear.
no code implementations • NAACL 2021 • Aina Garí Soler, Marianna Apidianaki
The intensity relationship that holds between scalar adjectives (e. g., nice < great < wonderful) is highly relevant for natural language inference and common-sense reasoning.
1 code implementation • 29 Apr 2021 • Aina Garí Soler, Marianna Apidianaki
Pre-trained language models (LMs) encode rich information about linguistic structure but their knowledge about lexical polysemy remains unclear.
1 code implementation • EMNLP 2020 • Aina Garí Soler, Marianna Apidianaki
Adjectives like pretty, beautiful and gorgeous describe positive properties of the nouns they modify but with different intensity.
1 code implementation • SEMEVAL 2020 • Aina Garí Soler, Marianna Apidianaki
Our best English models occupy the third and fourth positions in the ranking for the two subtasks.
no code implementations • JEPTALNRECITAL 2019 • Syrielle Montariol, Aina Garí Soler, Alexandre Allauzen
This study is a preliminary exploration of the concept of informativeness -how much information a sentence gives about a word it contains- and its potential benefits to building quality word representations from scarce data.
no code implementations • SEMEVAL 2019 • Aina Garí Soler, Marianna Apidianaki, Alexandre Allauzen
Usage similarity estimation addresses the semantic proximity of word instances in different contexts.