Enhancing Bilingual Lexicon Induction with Dynamic Translation
| Authors | |
|---|---|
| Year of publication | 2025 |
| Type | Article in Proceedings |
| Conference | Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART |
| MU Faculty or unit | |
| Citation | DENISOVÁ, Michaela and Pavel RYCHLÝ. Enhancing Bilingual Lexicon Induction with Dynamic Translation. Online. In Ana Paula Rocha, Luc Steels, H. Jaap van den Herik. Proceedings of the 17th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART. Porto: SciTePress, 2025, p. 735-744. ISBN 978-989-758-737-5. Available from: https://dx.doi.org/10.5220/0013346000003890. |
| Doi | http://dx.doi.org/10.5220/0013346000003890 |
| Keywords | cross-Lingual embedding models; parameter k; classification neural network |
| Description | Bilingual lexicon induction (BLI) has been a popular task for evaluating cross-lingual word embeddings (CWEs). The prevalent metric employed in the evaluation is precision at k, where k represents the number of target words retrieved for each source word. However, establishing a fixed k for the entire evaluation dataset proves challenging due to varying target word counts for each source word. This leads to limited results, compromising either precision or recall. In this paper, we present a novel classification-based approach with dynamic k for bilingual lexicon induction that aims to identify all relevant target words for each source word by exploiting the information derived from the aligned embeddings while offering a balanced trade-off between precision and recall. On top of that, it enables the evaluation of the existing CWEs using dynamic k. Compared to the standard baseline systems and evaluation procedures, it provides competitive results. |
| Related projects: |