Papers by Viggo Unmack Gascou
How to Encode Domain Information in Relation Classification (2024.lrec-main)
Copied to clipboard
Elisa Bassignana, Viggo Unmack Gascou, Frida Nøhr Laustsen, Gustav Kristensen, Marie Haahr Petersen, Rob van der Goot, Barbara Plank
| Challenge: | Existing deep learning models require a lot of training data to obtain high performance. |
| Approach: | They propose a multi-domain training setup for Relation Classification (RC) they compare different ways to enrich input instances with domain information . |
| Outcome: | The proposed model improves > 2 Macro-F1 against the baseline setup. |