Event Relation Extraction (ERE) aims to identify and classify semantic relationships between events expressed in text. While existing work has mainly addressed temporal or simple causal links, fine-grained causal relations such as enable, prevent, and intend remain insufficiently explored, partly due to limited and imbalanced labeled datasets. We present a novel framework that leverages large language models (LLMs) and common-sense knowledge to jointly perform event extraction and relation classification. Our contribution includes (1) the creation of the CausalSense large-scale dataset containing more than 500k sentences from news data and commonsense knowledge extracted from ATOMIC, and enriched synthetically; and (2) the evaluation of multiple architectures, including transformer-based models and end-to-end multitask systems for extracting fine-grained causal relationships. Experimental results show that our best-performing model achieves a 32.3% improvement in average F1-score over the current state of the art. The integration of commonsense knowledge substantially enhances fine-grained causal relation detection. The CausalSense dataset, our code and models are released as open source to support future research on causal event relationship extraction.

While prior work in Information Extraction (IE) has focused on extracting information from either textual content or tables in isolation, they miss critical information that emerges only from their interplay. Indeed, tables may summarize facts sparse in the text, while text can disambiguate or elaborate on table entries. This complementarity may take the form of relations which are expressed across text and tables. In this context, we are interested in the task of extracting such relations whose expression spans the two modalities. This task is an original one, for which no reference evaluation corpora exists. Thus we created ReTaT, a corpus that can be used to train and evaluate systems for extracting such relations. This corpus is composed of (table, surrounding text) pairs extracted from Wikipedia pages and has been manually annotated with relation triples. ReTaT is organized in three datasets with distinct characteristics: domain (business, telecommunication and female celebrities), size (from 50 to 255 pairs), language (English vs French), type of relations (data vs object properties), close vs open list of relation, size of the surrounding text (paragraph vs full page). We then assessed its quality and suitability for the joint table-text relation extraction task using Large Language Models (LLMs), at a time when LLMs have demonstrated their ability to extract relations from either text or tables in isolation.