Towards Fine-grained Causal Reasoning and QA
This addresses the need for improved causal understanding in high-stakes NLP applications, though it is incremental as it builds on existing causal reasoning work with new data and tasks.
The paper tackled the problem of fine-grained causal reasoning in NLP by introducing a novel dataset with 25K cause-effect event pairs and 24K QA pairs, showing that state-of-the-art methods face unique challenges across tasks like causality detection and Causal QA.
Understanding causality is key to the success of NLP applications, especially in high-stakes domains. Causality comes in various perspectives such as enable and prevent that, despite their importance, have been largely ignored in the literature. This paper introduces a novel fine-grained causal reasoning dataset and presents a series of novel predictive tasks in NLP, such as causality detection, event causality extraction, and Causal QA. Our dataset contains human annotations of 25K cause-effect event pairs and 24K question-answering pairs within multi-sentence samples, where each can have multiple causal relationships. Through extensive experiments and analysis, we show that the complex relations in our dataset bring unique challenges to state-of-the-art methods across all three tasks and highlight potential research opportunities, especially in developing "causal-thinking" methods.