Tutorial

Title: Counterspeech Against Hate and Misinformation: An NLP Perspective

Abstract: This tutorial bridges computer science, social sciences, and public policy to examine how counterspeech is deployed by individuals, activists, and organizations to combat online abuse and misinformation. Specifically, we focus on how Natural Language Processing (NLP) and Generation (NLG) can automate counterspeech production, analyzing both the technical breakthroughs and the ethical implications of leveraging Large Language Models (LLMs) for this task. Beyond generation, the tutorial addresses critical socio-technical dimensions, including the development of robust evaluation metrics to measure real-world impact, the vital role of domain experts in crafting high-quality datasets and taxonomies, and strategies to mitigate inherent biases in language models. Finally, we delve into the complex, frontier challenge of tackling hate speech and misinformation simultaneously—an open research question that is rapidly gaining standalone traction in the NLP community.

---

Short bios:

Marco Guerini holds a bachelor’s degree in Philosophy and earned his Ph.D. in Information and Communication Technologies from the University of Trento. He is currently a researcher in Computational Linguistics and head of the Language and Dialogue Technologies (LanD) group at the Fondazione Bruno Kessler (FBK). His work focuses on persuasive communication, sentiment analysis, and social media. In recent years, his research has centered on developing artificial intelligence technologies to support the generation of counterspeech aimed at combating online hate speech and misinformation.

Daniel Russo is a researcher at Fondazione Bruno Kessler (FBK), where he is a member of the Language and Dialogue Technologies group. He completed his PhD in Information Engineering and Computer Science at the University of Trento, in collaboration with FBK, under the supervision of Marco Guerini and Jacopo Staiano. He holds an MSc in Cognitive Science and a BSc in Computer Science. His research centers on Natural Language Processing and Generation for countering online misinformation and hate speech, spanning automated fact-checking and counterspeech generation.