Interpretability is a topic that has been in the spotlight for the past few years. Most existing interpretability techniques produce interpretations in the form of rules or feature importance. These interpretations, while informative, may be harder to understand for non-expert users and therefore, cannot always be considered as adequate explanations. To that end, explanations in natural language are often preferred, as they are easier to comprehend and also more presentable to end-users. This work introduces an early concept for a novel pipeline that can be used in text classification tasks, offering predictions and explanations in natural language. It comprises of two models: a classifier for labelling the text and an explanation generator which provides the explanation. The proposed pipeline can be adopted by any text classification task, given that ground truth rationales are available to train the explanation generator. Our experiments are centred around the tasks of sentiment analysis and offensive language identification in Greek tweets, using a Greek Large Language Model (LLM) to obtain the necessary explanations that can act as rationales. The experimental evaluation was performed through a user study based on three different metrics and achieved promising results for both datasets.

The prevalence of offensive content on the internet, encompassing hate speech and cyberbullying, is a pervasive issue worldwide. Consequently, it has garnered significant attention from the machine learning (ML) and natural language processing (NLP) communities. As a result, numerous systems have been developed to automatically identify potentially harmful content and to mitigate its impact. These systems can follow two approaches; (i) Use publicly available models and application endpoints, including prompting large language models (LLMs) (ii) Annotate datasets and train ML models on them. However, both approaches lack an understanding of how generalizable they are. Furthermore, the applicability of these systems is often questioned in off-domain and practical environments. This paper empirically evaluates the generalizability of offensive language detection models and datasets across a novel generalized benchmark: GenOffense. We answer three research questions on generalizability. Our findings will be useful in creating robust real-world offensive language detection systems.

The spread of various forms of offensive speech online is an important concern in social media. While platforms have been investing heavily in ways of coping with this problem, the question of privacy remains largely unaddressed. Models trained to detect offensive language on social media are trained and/or fine-tuned using large amounts of data often stored in centralized servers. Since most social media data originates from end users, we propose a privacy preserving decentralized architecture for identifying offensive language online by introducing Federated Learning (FL) in the context of offensive language identification. FL is a decentralized architecture that allows multiple models to be trained locally without the need for data sharing hence preserving users' privacy. We propose a model fusion approach to perform FL. We trained multiple deep learning models on four publicly available English benchmark datasets (AHSD, HASOC, HateXplain, OLID) and evaluated their performance in detail. We also present initial cross-lingual experiments in English and Spanish. We show that the proposed model fusion approach outperforms baselines in all the datasets while preserving privacy.

The ubiquity of offensive content on social media is a growing cause for concern among companies and government organizations. Recently, transformer-based models such as BERT, XLNET, and XLM-R have achieved state-of-the-art performance in detecting various forms of offensive content (e.g. hate speech, cyberbullying, and cyberaggression). However, the majority of these models are limited in their capabilities due to their encoder-only architecture, which restricts the number and types of labels in downstream tasks. Addressing these limitations, this study presents the first pre-trained model with encoder-decoder architecture for offensive language identification with text-to-text transformers (T5) trained on two large offensive language identification datasets; SOLID and CCTK. We investigate the effectiveness of combining two datasets and selecting an optimal threshold in semi-supervised instances in SOLID in the T5 retraining step. Our pre-trained T5 model outperforms other transformer-based models fine-tuned for offensive language detection, such as fBERT and HateBERT, in multiple English benchmarks. Following a similar approach, we also train the first multilingual pre-trained model for offensive language identification using mT5 and evaluate its performance on a set of six different languages (German, Hindi, Korean, Marathi, Sinhala, and Spanish). The results demonstrate that this multilingual model achieves a new state-of-the-art on all the above datasets, showing its usefulness in multilingual scenarios. Our proposed T5-based models will be made freely available to the community.

Identifying offensive content in social media is vital for creating safe online communities. Several recent studies have addressed this problem by creating datasets for various languages. In this paper, we explore offensive language identification in texts with transliterations and code-mixing, linguistic phenomena common in multilingual societies, and a known challenge for NLP systems. We introduce TB-OLID, a transliterated Bangla offensive language dataset containing 5,000 manually annotated comments. We train and fine-tune machine learning models on TB-OLID, and we evaluate their results on this dataset. Our results show that English pre-trained transformer-based models, such as fBERT and HateBERT achieve the best performance on this dataset.

Code-mixing is a well-studied linguistic phenomenon when two or more languages are mixed in text or speech. Several works have been conducted on building datasets and performing downstream NLP tasks on code-mixed data. Although it is not uncommon to observe code-mixing of three or more languages, most available datasets in this domain contain code-mixed data from only two languages. In this paper, we introduce OffMix-3L, a novel offensive language identification dataset containing code-mixed data from three different languages. We experiment with several models on this dataset and observe that BanglishBERT outperforms other transformer-based models and GPT-3.5.

Text classification is an area of research which has been studied over the years in Natural Language Processing (NLP). Adapting NLP to multiple domains has introduced many new challenges for text classification and one of them is long document classification. While state-of-the-art transformer models provide excellent results in text classification, most of them have limitations in the maximum sequence length of the input sequence. The majority of the transformer models are limited to 512 tokens, and therefore, they struggle with long document classification problems. In this research, we explore on employing Model Fusing for long document classification while comparing the results with well-known BERT and Longformer architectures.

The widespread of offensive content online, such as hate speech and cyber-bullying, is a global phenomenon. This has sparked interest in the artificial intelligence (AI) and natural language processing (NLP) communities, motivating the development of various systems trained to detect potentially harmful content automatically. These systems require annotated datasets to train the machine learning (ML) models. However, with a few notable exceptions, most datasets on this topic have dealt with English and a few other high-resource languages. As a result, the research in offensive language identification has been limited to these languages. This paper addresses this gap by tackling offensive language identification in Sinhala, a low-resource Indo-Aryan language spoken by over 17 million people in Sri Lanka. We introduce the Sinhala Offensive Language Dataset (SOLD) and present multiple experiments on this dataset. SOLD is a manually annotated dataset containing 10,000 posts from Twitter annotated as offensive and not offensive at both sentence-level and token-level, improving the explainability of the ML models. SOLD is the first large publicly available offensive language dataset compiled for Sinhala. We also introduce SemiSOLD, a larger dataset containing more than 145,000 Sinhala tweets, annotated following a semi-supervised approach.

The presence of offensive language on social media is very common motivating platforms to invest in strategies to make communities safer. This includes developing robust machine learning systems capable of recognizing offensive content online. Apart from a few notable exceptions, most research on automatic offensive language identification has dealt with English and a few other high resource languages such as French, German, and Spanish. In this paper we address this gap by tackling offensive language identification in Marathi, a low-resource Indo-Aryan language spoken in India. We introduce the Marathi Offensive Language Dataset v.2.0 or MOLD 2.0 and present multiple experiments on this dataset. MOLD 2.0 is a much larger version of MOLD with expanded annotation to the levels B (type) and C (target) of the popular OLID taxonomy. MOLD 2.0 is the first hierarchical offensive language dataset compiled for Marathi, thus opening new avenues for research in low-resource Indo-Aryan languages. Finally, we also introduce SeMOLD, a larger dataset annotated following the semi-supervised methods presented in SOLID.

The widespread of offensive content online has become a reason for great concern in recent years, motivating researchers to develop robust systems capable of identifying such content automatically. With the goal of carrying out a fair evaluation of these systems, several international competitions have been organized, providing the community with important benchmark data and evaluation methods for various languages. Organized since 2019, the HASOC (Hate Speech and Offensive Content Identification) shared task is one of these initiatives. In its fourth iteration, HASOC 2022 included three subtracks for English, Hindi, and Marathi. In this paper, we report the results of the HASOC 2022 Marathi subtrack which provided participants with a dataset containing data from Twitter manually annotated using the popular OLID taxonomy. The Marathi track featured three additional subtracks, each corresponding to one level of the taxonomy: Task A - offensive content identification (offensive vs. non-offensive); Task B - categorization of offensive types (targeted vs. untargeted), and Task C - offensive target identification (individual vs. group vs. others). Overall, 59 runs were submitted by 10 teams. The best systems obtained an F1 of 0.9745 for Subtrack 3A, an F1 of 0.9207 for Subtrack 3B, and F1 of 0.9607 for Subtrack 3C. The best performing algorithms were a mixture of traditional and deep learning approaches.