Although there is an unprecedented effort to provide adequate responses in terms of laws and policies to hate content on social media platforms, dealing with hatred online is still a tough problem. Tackling hate speech in the standard way of content deletion or user suspension may be charged with censorship and overblocking. One alternate strategy, that has received little attention so far by the research community, is to actually oppose hate content with counter-narratives (i.e. informed textual responses). In this paper, we describe the creation of the first large-scale, multilingual, expert-based dataset of hate speech/counternarrative pairs. This dataset has been built with the effort of more than 100 operators from three different NGOs that applied their training and expertise to the task. Together with the collected data we also provide additional annotations about expert demographics, hate and response type, and data augmentation through translation and paraphrasing. Finally, we provide initial experiments to assess the quality of our data.

Several NLP studies address the problem of figurative language, but among non-literal phenomena, they have neglected exaggeration. This paper presents a first computational approach to this figure of speech. We explore the possibility to automatically detect exaggerated sentences. First, we introduce HYPO, a corpus containing overstatements (or hyperboles) collected on the web and validated via crowdsourcing. Then, we evaluate a number of models trained on HYPO, and bring evidence that the task of hyperbole identification can be successfully performed based on a small set of semantic features.

Recently research has started focusing on avoiding undesired effects that come with content moderation, such as censorship and overblocking, when dealing with hatred online. The core idea is to directly intervene in the discussion with textual responses that are meant to counter the hate content and prevent it from further spreading. Accordingly, automation strategies, such as natural language generation, are beginning to be investigated. Still, they suffer from the lack of sufficient amount of quality data and tend to produce generic/repetitive responses. Being aware of the aforementioned limitations, we present a study on how to collect responses to hate effectively, employing large scale unsupervised language models such as GPT-2 for the generation of silver data, and the best annotation strategies/neural architectures that can be used for data filtering before expert validation/postediting.

Language is the main communication device to represent the environment and share a common understanding of the world that we perceive through our sensory organs. Therefore, each language might contain a great amount of sensorial elements to express the perceptions both in literal and figurative usage. To tackle the semantics of figurative language, several conceptual properties such as concreteness or imegeability are utilized. However, there is no attempt in the literature to analyze and benefit from the sensorial elements for figurative language processing. In this paper, we investigate the impact of sensorial features on metaphor identification. We utilize an existing lexicon associating English words to sensorial modalities and propose a novel technique to automatically discover these associations from a dependency-parsed corpus. In our experiments, we measure the contribution of the sensorial features to the metaphor identification task with respect to a state of the art model. The results demonstrate that sensorial features yield better performance and show good generalization properties.

Connecting words with senses, namely, sight, hearing, taste, smell and touch, to comprehend the sensorial information in language is a straightforward task for humans by using commonsense knowledge. With this in mind, a lexicon associating words with senses would be crucial for the computational tasks aiming at interpretation of language. However, to the best of our knowledge, there is no systematic attempt in the literature to build such a resource. In this paper, we present a sensorial lexicon that associates English words with senses. To obtain this resource, we apply a computational method based on bootstrapping and corpus statistics. The quality of the resulting lexicon is evaluated with a gold standard created via crowdsourcing. The results show that a simple classifier relying on the lexicon outperforms two baselines on a sensory classification task, both at word and sentence level, and confirm the soundness of the proposed approach for the construction of the lexicon and the usefulness of the resource for computational applications.