TartuNLP @ AXOLOTL-24: Leveraging Classifier Output for New Sense Detection in Lexical Semantics

TL;DR

This paper describes a simple, efficient approach using adapter-based classifiers to detect and define new word senses over time, achieving top results in a shared task.

Contribution

The authors introduce a novel, computationally inexpensive method leveraging classifier probabilities for new sense detection and definition generation.

Findings

Achieved third place in new sense detection.

Secured first place in sense definition generation.

Proved effectiveness of adapter-based models in lexical semantics.

Abstract

We present our submission to the AXOLOTL-24 shared task. The shared task comprises two subtasks: identifying new senses that words gain with time (when comparing newer and older time periods) and producing the definitions for the identified new senses. We implemented a conceptually simple and computationally inexpensive solution to both subtasks. We trained adapter-based binary classification models to match glosses with usage examples and leveraged the probability output of the models to identify novel senses. The same models were used to match examples of novel sense usages with Wiktionary definitions. Our submission attained third place on the first subtask and the first place on the second subtask.