Leveraging Declarative Knowledge in Text and First-Order Logic for Fine-Grained Propaganda Detection

Abstract

We study the detection of propagandistic text fragments in news articles. Instead of merely learning from input-output datapoints in training data, we introduce an approach to inject declarative knowledge of fine-grained propaganda techniques. Specifically, we leverage the declarative knowledge expressed in both first-order logic and natural language. The former refers to the logical consistency between coarse-and fine-grained predictions, which is used to regularize the training process with propositional Boolean expressions. The latter refers to the literal definition of each propaganda technique, which is utilized to get class representations for regularizing the model parameters. We conduct experiments on Propaganda Techniques Corpus, a large manually annotated dataset for fine-grained propaganda detection. Experiments show that our method achieves superior performance, demonstrating that leveraging declarative knowledge can help the model to make more accurate predictions.

Mitigating Reporting Bias in Semi-supervised Temporal Commonsense Inference with Probabilistic Soft Logic

“…The soft probabilistic logic combines logic's expressive power with the ability to deal with uncertainty, which has been introduced in a variety of reasoning tasks ranging from Knowledge Base Completion (Yang, Yang, and Cohen 2017;Chen et al 2019;Mohler, Monahan, and Tomlinson 2020) and Social Prediction (Wang et al 2020) to Temporal Relation Extraction (Zhou et al 2021b) and Causal Inference (Sridhar and Getoor 2016;Du et al 2021). Most of the works inject the logic knowledge to neural networks by introducing logic-driven loss functions.…”