Lacramioara Dranca


2023

pdf
No clues good clues: out of context Lexical Relation Classification
Lucia Pitarch | Jordi Bernad | Lacramioara Dranca | Carlos Bobed Lisbona | Jorge Gracia
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

The accurate prediction of lexical relations between words is a challenging task in Natural Language Processing (NLP). The most recent advances in this direction come with the use of pre-trained language models (PTLMs). A PTLM typically needs “well-formed” verbalized text to interact with it, either to fine-tune it or to exploit it. However, there are indications that commonly used PTLMs already encode enough linguistic knowledge to allow the use of minimal (or none) textual context for some linguistically motivated tasks, thus notably reducing human effort, the need for data pre-processing, and favoring techniques that are language neutral since do not rely on syntactic structures. In this work, we explore this idea for the tasks of lexical relation classification (LRC) and graded Lexical Entailment (LE). After fine-tuning PTLMs for LRC with different verbalizations, our evaluation results show that very simple prompts are competitive for LRC and significantly outperform graded LE SoTA. In order to gain a better insight into this phenomenon, we perform a number of quantitative statistical analyses on the results, as well as a qualitative visual exploration based on embedding projections.

2020

pdf
Multi-Strategy system for translation inference across dictionaries
Lacramioara Dranca
Proceedings of the 2020 Globalex Workshop on Linked Lexicography

This paper describes four different strategies proposed to the TIAD 2020 Shared Task for automatic translation inference across dictionaries. The proposed strategies are based on the analysis of Apertium RDF graph, taking advantage of characteristics such as translation using multiple paths, synonyms and similarities between lexical entries from different lexicons and cardinality of possible translations through the graph. The four strategies were trained and validated on the Apertium RDF EN<->ES dictionary, showing promising results. Finally, the strategies, applied together, obtained an F-measure of 0.43 in the task of inferring the dictionaries proposed in the shared task, ranking thus third with respect to the other new systems presented to the TIAD 2020 Shared Task. No system presented to the shared task exceeded the baseline proposed by the TIAD organizers.