In the absence of sense-annotated data, word sense induction (WSI) is a compelling alternative to word sense disambiguation, particularly in low-resource or domain-specific settings. In this paper, we emphasize methodological problems in current WSI evaluation. We propose an evaluation on a SemCor-derived dataset, respecting the original corpus polysemy and frequency distributions. We assess pre-trained embeddings and clustering algorithms across parts of speech, and propose and evaluate an LLM-based WSI method for English. We evaluate data augmentation sources (LLM-generated, corpus and lexicon), and semi-supervised scenarios using Wiktionary for data augmentation, must-link constraints, number of clusters per lemma.We find that no unsupervised method (whether ours or previous) surpasses the strong “one cluster per lemma” heuristic (1cpl). We also show that (i) results and best systems may vary across POS, (ii) LLMs have troubles performing this task, (iii) data augmentation is beneficial and (iv) capitalizing on Wiktionary does help. It surpasses previous SOTA system on our test set by 3.3%. WSI is not solved, and calls for a better articulation of lexicons and LLMs’ lexical semantics capabilities.

Les modèles de langue pré-entraînés ont apporté des avancées significatives dans les représentations contextuelles des phrases et des mots. Cependant, les tâches lexicales restent un défi pour ces représentations en raison des problèmes tels que la faible similarité des representations d’un même mot dans des contextes similaires. Mosolova et al. (2024) ont montré que l’apprentissage contrastif supervisé au niveau des tokens permettait d’améliorer les performances sur les tâches lexicales. Dans cet article, nous étudions la généralisabilité de leurs résultats obtenus en anglais au français, à d’autres modèles de langue et à plusieurs parties du discours. Nous démontrons que cette méthode d’apprentissage contrastif améliore systématiquement la performance sur les tâches de Word-in-Context et surpasse celle des modèles de langage pré-entraînés standards. L’analyse de l’espace des plongements lexicaux montre que l’affinage des modèles rapproche les exemples ayant le même sens et éloigne ceux avec des sens différents, ce qui indique une meilleure discrimination des sens dans l’espace vectoriel final.

While static word embeddings are blind to context, for lexical semantics tasks context is rather too present in contextual word embeddings, vectors of same-meaning occurrences being too different (Ethayarajh, 2019). Fine-tuning pre-trained language models (PLMs) using contrastive learning was proposed, leveraging automatically self-augmented examples (Liu et al., 2021b). In this paper, we investigate how to inject a lexicon as an alternative source of supervision, using the English Wiktionary. We also test how dimensionality reduction impacts the resulting contextual word embeddings. We evaluate our approach on the Word-In-Context (WiC) task, in the unsupervised setting (not using the training set). We achieve new SoTA result on the original WiC test set. We also propose two new WiC test sets for which we show that our fine-tuning method achieves substantial improvements. We also observe improvements, although modest, for the semantic frame induction task. Although we experimented on English to allow comparison with related work, our method is adaptable to the many languages for which large Wiktionaries exist.

Numerous machine translation systems have been proposed since the appearance of this task. Nowadays, new large language model-based algorithms show results that sometimes overcome human ones on the rich-resource languages. Nevertheless, it is still not the case for the low-resource languages, for which all these algorithms did not show equally impressive results. In this work, we want to compare 3 generations of machine translation models on 7 low-resource languages and make a step further by proposing a new way of automatic parallel data augmentation using the state-of-the-art generative model.

We present an algorithm for detecting metaphor in sentences which was used in Shared Task on Metaphor Detection by First Workshop on Figurative Language Processing. The algorithm is based on different features and Conditional Random Fields.