Deep Learning

MAFALDA is a new benchmark for fallacy classification that consolidates previous datasets into a unified taxonomy. The benchmark includes manual annotations with explanations, a specialized annotation scheme for subjective NLP tasks, and a novel evaluation method to handle subjectivity. We assessed both language models and human performance on MAFALDA to evaluate their fallacy detection and classification capabilities.

The NoRDF Project is a scientific project at Telecom Paris that aims to model and extract complex information from natural language text. More precisely, we want to enrich knowledge bases with events, causation, conditions, precedence, stories, negation, and beliefs.