| # Webis Clickbait Spoiling Corpus | |
| The Webis Clickbait Spoiling Corpus 2022 (Webis-Clickbait-22) contains 5,000 spoiled clickbait posts crawled from Facebook, Reddit, and Twitter. | |
| This corpus supports the task of clickbait spoiling, which deals with generating a short text that satisfies the curiosity induced by a clickbait post. | |
| This dataset contains the clickbait posts and manually cleaned versions of the linked documents, and extracted spoilers for each clickbait post. | |
| Additionally, the spoilers are categorized into three types: short phrase spoilers, longer passage spoilers, and multiple non-consecutive pieces of text. The test set of this dataset was used for the [SemEval-2023 clickbait spoiling task](https://www.tira.io/task-overview/clickbait-spoiling). You can re-execute and adopt the software submissions made through for this SemEval task, please see the instructions and overview of approaches in [TIRA](https://www.tira.io/task-overview/clickbait-spoiling). | |
| ## Overview | |
| The dataset comes with predefined train/validation/test splits: | |
| - [3,200 posts for training](training.jsonl) | |
| - [800 posts for validation](validation.jsonl) | |
| - [1,000 posts for testing](test.jsonl) | |
| - The test set was used for the [SemEval-2023 clickbait spoiling task](https://www.tira.io/task-overview/clickbait-spoiling). This shared task was organized with [TIRA.io](https://www.tira.io/) and participants submitted Docker software during the task. Please see the instructions in [TIRA](https://www.tira.io/task-overview/clickbait-spoiling) to re-execute or modify the approaches. | |