Shrome at Touché: Soft-Vote Ensembling and Counter-Causal Augmentation for Causality Extraction

TL;DR

This paper proposes a system for causality extraction, including counter-causal claims. The system uses soft-vote ensembling and counter-causal augmentation to improve performance. It achieves state-of-the-art results on the Countercausal News Corpus.

Why It Matters

Accurate causality extraction is crucial for natural language understanding and has applications in areas like fact-checking and decision-making. The system's ability to handle counter-causal claims enhances its practical significance. This research contributes to the development of more robust language models.

Key Findings

• Soft-vote ensembling improves extraction performance
• Counter-causal augmentation enhances polarity labeling
• The system achieves state-of-the-art results on the Countercausal News Corpus