Valerio Basile

Institution: University of Turín (Italy).

Title: How to Model the Voice of the Minority: a Perspectivist Approach to NLP

Abstract: Contemporary Natural Language Processing is largely rooted in language resources, e.g., for training models in supervised machine learning. Even is few- or zero-shot settings, the importance of good quality data for benchmarking is paramount. Typically, data come in the form of large manually annotated datasets. The harmonization of the annotation, however, is often problematic, especially when highly subjective annotation tasks are performed, such as sentiment, irony, undesirable language, and generally anything involving pragmatics. Discarding and averaging discordant opinions carries the risk of losing the rich knowledge coming from different annotators’ perspectives.
A recent line of research proposes to never aggregate annotations [1], but rather to leverage the worth of knowledge found in disagreement for building models [2] and evaluating them [3].
In this talk, I will present the perspectivist paradigm, focusing on NLP, and a few works exploring its implications on data annotation, model evaluation, and interoperability, with particular focus on the perspectivist modeling of hate speech and irony.

[1] The Perspectivist Manifesto. https://pdai.info/
[2] Basile et al. 2021 “We Need to Consider Disagreement in Evaluation” https://aclanthology.org/2021.bppf-1.3/
[3] Cabitza et al. “Toward a perspectivist turn in ground truthing for predictive computing”. AAAI-23. https://arxiv.org/pdf/2109.04270.pdf

Bio: Valerio Basile is an Assistant Professor at the Computer Science Department of the University of Turin, Italy, member of the Content-centered Computing group and the Hate Speech Monitoring group. His work spans across several areas such as: formal representations of meaning, linguistic annotation, natural language generation, commonsense knowledge, semantic parsing, sentiment analysis, and hate speech detection, perspectives and bias in supervised machine learning, from data creation to system evaluation. He is currently PI of the project BREAKhateDOWN “Toxic Language Understanding in Online Communication”, and among the main proponents of the Perspectivist Data Manifesto: https://pdai.info

Elena Cabrio

Institution: University of Côte d’Azur (Francia).

Title: Processing Natural Language to Extract, Analyze and Generate Arguments from Texts.

Abstract: Intelligent machines enriched with computational argumentation models can extract, analyse,  summarise and generate natural language argumentative structures from different contexts and  from various textual resources. In this talk, I will first introduce the research area of Argument Mining,  addressing the challenge of identifying argumentative components and structures from texts. Then, I will discuss how these methods can be used in two different scenarios: 1) on medical texts: to  enhance evidence-based medicine with natural language argumentative analysis of clinical trials and  to generate argument-based natural language explanations for the correct and incorrect answers of standardized  medical exams, and 2) on political texts: to identify fallacious arguments in political debates. I will then conclude  with some  thoughts on  the automatic generation of counter-arguments to fight online disinformation  and hate speech.

Bio: Elena Cabrio is Full Professor at the University of Côte d’Azur and member of the Inria-I3S research team Wimmics. In 2021 she was awarded with a Chair in Artificial Intelligence at the Interdisciplinary Institute for Artificial Intelligence 3IA Côte d’Azur on “AI and Natural Language”. Her main research interests are in NLP, mainly Argumentation Mining, Information Extraction and Hate Speech detection. Goal of her research is to design debating technologies for advanced decision support systems, to support the exchange of information and opinions in different domains (as healthcare and politics), leveraging interdisciplinarity and advances in machine learning for Natural Language Processing. She has published over 100 scientific articles, including journals and top conferences in AI and NLP.  She is currently coordinating the ANTIDOTE (ArgumeNtaTIon-Driven explainable artificial intelligence fOr digiTal mEdicine) project (CHIST-ERA XAI 2019).