Extension usage examples:
Here's how our browser extension sees the article:
Article summary:
1. The article presents a large-scale stance detection dataset extracted from comments written by candidates of elections in Switzerland.
2. The dataset consists of German, French and Italian text, allowing for a cross-lingual evaluation of stance detection.
3. A single model is used to train on all the issues, with prepended natural questions representing the target (e.g. "Do you support X?").
Article analysis:
The article is reliable and trustworthy as it provides detailed information about the dataset and its purpose, as well as baseline results from multilingual BERT that show that zero-shot cross-lingual and cross-target transfer of stance detection is moderately successful with this approach. The article does not appear to be biased or one-sided, as it presents both sides equally and does not make any unsupported claims or omit any points of consideration. Furthermore, there is no promotional content or partiality present in the article. The article does note possible risks associated with using the dataset for stance detection, such as potential errors due to language differences or incorrect labeling of data points. All in all, the article appears to be reliable and trustworthy overall.