Summary of Austrotox: a Dataset For Target-based Austrian German Offensive Language Detection, by Pia Pachinger et al.
AustroTox: A Dataset for Target-Based Austrian German Offensive Language Detection
by Pia Pachinger, Janis Goldzycher, Anna Maria Planitzer, Wojciech Kusa, Allan Hanbury, Julia Neidhardt
First submitted to arxiv on: 12 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Computers and Society (cs.CY)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a novel dataset for toxicity detection in online forums, with a focus on the Austrian German dialect. The dataset comprises 4,562 user comments annotated at the token level to identify offensive language, including vulgar language and targets of offensive statements. The authors evaluate fine-tuned language models as well as large language models in zero- and few-shot settings, finding that while fine-tuned models excel in detecting linguistic peculiarities, large language models demonstrate superior performance in detecting offensiveness. The paper’s contributions include the introduction of a new dataset and the evaluation of various language model architectures for toxicity detection. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a special dataset to help computers understand when people are saying something mean or rude online. They used comments from an Austrian news forum, where people often speak with a different accent. The dataset has lots of examples, over 4,500 in total! The authors tested different computer programs that can understand language and found out which ones work best for detecting mean things. They’re sharing the data and code so other researchers can use it too. |
Keywords
» Artificial intelligence » Few shot » Language model » Token
