Event Causality Identification via Generation of Important Context Words

Abstract

An important problem of Information Extraction involves Event Causality Identification (ECI) that seeks to identify causal relation between pairs of event mentions. Prior models for ECI have mainly solved the problem using the classification framework that does not explore prediction/generation of important context words from input sentences for causal recognition. In this work, we consider the words along the dependency path between the two event mentions in the dependency tree as the important context words for ECI. We introduce dependency path generation as a complementary task for ECI, which can be solved jointly with causal label prediction to improve the performance. To facilitate the multi-task learning, we cast ECI into a generation problem that aims to generate both causal relation and dependency path words from input sentence. In addition, we propose to use the REIN-FORCE algorithm to train our generative model where novel reward functions are designed to capture both causal prediction accuracy and generation quality. The experiments on two benchmark datasets demonstrate state-of-theart performance of the proposed model for ECI.

Mastering Context-to-Label Representation Transformation for Event Causality Identification with Diffusion Models

“…In addition to the use of pre-trained language models, at the core of such deep learning models characterizes different additional resources to advance the performance for ECI, such as distant supervision data (Zuo et al 2020), background knowledge (Liu, Chen, and Zhao 2020), dependency parsing (Tran Phu and Nguyen 2021), and external causal statements (Zuo et al 2021a). Recently, some work has also explored a new formulation for ECI using generative models to demonstrate promising performance (Man, Nguyen, and Nguyen 2022;Shen et al 2022). However, the common issue of previous ECI models concerns the drastic changes in the processes to transform event context representations to causal label representations, which cannot secure optimal performance.…”

IDIAPers @ Causal News Corpus 2022: Efficient Causal Relation Identification Through a Prompt-based Few-shot Approach

“…Once the causal information is identified within a 1 Code available at https://github.com/idiap/cncsharedtask. text, such knowledge becomes beneficial for many other downstream NLP tasks, e.g., Information Extraction, Question Answering, Text Summarization (Ayyanar et al, 2019a;Man et al, 2022). However, due to the ambiguity and diversity in written documents, causality identification is not easy and remains a challenging problem.…”

Deep neural networks for identifying causal relations in texts

“…Event Causality Identification (ECI) plays a vital role in many downstream applications such as why-question answering, logical reasoning, and event prediction, etc. The task of ECI is usually formulated as the event pair classification problem [86][87][88][89], where the event pair classifier is built fully based on the training data.…”