Offensive language such as hate, abuse, and profanity (HAP) occurs in various content on the web. While previous work has mostly dealt with sentence level annotations, there have been a few recent attempts to identify offensive spans as well. We build upon this work and introduce Muted, a system to identify multilingual HAP content by displaying offensive arguments and their targets using heat maps to indicate their intensity. Muted can leverage any transformer-based HAP-classification model and its attention mechanism out-of-the-box to identify toxic spans, without further fine-tuning. In addition, we use the spaCy library to identify the specific targets and arguments for the words predicted by the attention heatmaps. We present the model's performance on identifying offensive spans and their targets in existing datasets and present new annotations on German text. Finally, we demonstrate our proposed visualization tool on multilingual inputs.

The rise in hateful and offensive language directed at other users is one of the adverse side effects of the increased use of social networking platforms. This could make it difficult for human moderators to review tagged comments filtered by classification systems. To help address this issue, we present the ViHOS (Vietnamese Hate and Offensive Spans) dataset, the first human-annotated corpus containing 26k spans on 11k comments. We also provide definitions of hateful and offensive spans in Vietnamese comments as well as detailed annotation guidelines. Besides, we conduct experiments with various state-of-the-art models. Specifically, XLM-R$_{Large}$ achieved the best F1-scores in Single span detection and All spans detection, while PhoBERT$_{Large}$ obtained the highest in Multiple spans detection. Finally, our error analysis demonstrates the difficulties in detecting specific types of spans in our data for future research. Disclaimer: This paper contains real comments that could be considered profane, offensive, or abusive.

(Sivanantham and Seran, 2019). It is widely spoken in the southern state of Tamil Nadu in India, Combating offensive content is crucial for different Sri Lanka, Malaysia, and Singapore. Tamil is an entities involved in content moderation, which official language of Tamil Nadu, Sri Lanka, Singapore, includes social media companies as well as individuals and the Union Territory of Puducherry in (Kumaresan et al., 2021; Chakravarthi and India. Significant minority speak Tamil in the four Muralidaran, 2021). To this end, moderation is other South Indian states of Kerala, Karnataka, often restrictive with either usage of human content Andhra Pradesh, and Telangana, as well as the moderators, who are expected to read through Union Territory of the Andaman and Nicobar Islands the content and flag the offensive mentions (Arsht (Sakuntharaj and Mahesan, 2021, 2017, 2016; and Etcovitch, 2018). Alternatively, there are Thavareesan and Mahesan, 2019, 2020a,b, 2021).

The interest in offensive content identification in social media has grown substantially in recent years. Previous work has dealt mostly with post level annotations. However, identifying offensive spans is useful in many ways. To help coping with this important challenge, we present MUDES, a multilingual system to detect offensive spans in texts. MUDES features pre-trained models, a Python API for developers, and a user-friendly web-based interface. A detailed description of MUDES' components is presented in this paper.