arxiv:2103.13606

Predicting Directionality in Causal Relations in Text

Published on Mar 25, 2021

Authors:

Abstract

Bidirectional transformer language models BERT and SpanBERT were evaluated for causal relation directionality prediction, with SpanBERT showing superior performance on longer spans and a unified framework for causal relation datasets introduced.

Generated by Qwen/Qwen2.5-Coder-32B-Instruct

In this work, we test the performance of two bidirectional transformer-based language models, BERT and SpanBERT, on predicting directionality in causal pairs in the textual content. Our preliminary results show that predicting direction for inter-sentence and implicit causal relations is more challenging. And, SpanBERT performs better than BERT on causal samples with longer span length. We also introduce CREST which is a framework for unifying a collection of scattered datasets of causal relations.