Skip to content

Latest commit

 

History

History
14 lines (10 loc) · 989 Bytes

README.md

File metadata and controls

14 lines (10 loc) · 989 Bytes
Raw

Few-Shot Learning for Argument Aspects of the Nuclear Energy Debate

This repository contains the experiment code for the paper:

Jurkschat, L., Wiedemann, G., Heinrich, M., Ruckdeschel, M., & Torge, S. (2022). Few-Shot Learning for Argument Aspects of the Nuclear Energy Debate. In Proceedings of the 13th International Conference on Language Resources and Evaluation (LREC 2022). European Language Resources Association (ELRA).

Experiments:

  1. Evaluation of different pre-trained transformer models to classify the AAC-NE dataset [1]
  2. Evaluation of two few-shot learning approaches to argument aspect mining
  3. Application of the best few-shot learner to a newspaper corpus of argumentative sentences from "The Guardian"

[1] Jurkschat, L., Wiedemann, G., Heinrich, M., Ruckdeschel, M., & Torge, S. (2022). Argument Aspect Corpus - Nuclear Energy. https://doi.org/10.5281/zenodo.6470232