# Clickbait Spoiling via Question Answering and Passage Retrieval

Source: [Clickbait Spoiling via Question Answering and Passage Retrieval](<https://www.sota2.com/research/paper/68fa6c141e209131cea31ffd>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-01T14:16:02.681Z

Authors: Matthias Hagen, Maik Fr\\"obe, Artur Jurk, Martin Potthast

Published: 2022-03-19T09:40:33.000Z

## Abstract

We introduce and study the task of clickbait spoiling: generating a short text that satisfies the curiosity induced by a clickbait post. Clickbait links to a web page and advertises its contents by arousing curiosity instead of providing an informative summary. Our contributions are approaches to classify the type of spoiler needed (i.e., a phrase or a passage), and to generate appropriate spoilers. A large-scale evaluation and error analysis on a new corpus of 5,000 manually spoiled clickbait posts -- the Webis Clickbait Spoiling Corpus 2022 -- shows that our spoiler type classifier achieves an accuracy of 80%, while the question answering model DeBERTa-large outperforms all others in generating spoilers for both types.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Clickbait Spoiling on Clickbait Posts Phrase Spoilers n=423 (test)](<https://www.sota2.com/research/sota/clickbait-spoiling-on-clickbait-posts-phrase-spoilers-n-423-test.md>): Clickbait Spoiling; Clickbait Posts Phrase Spoilers n=423 (test); BLEU-4: not recorded.

- [Clickbait Spoiling on Clickbait Posts Passage Spoilers n=403 (test)](<https://www.sota2.com/research/sota/clickbait-spoiling-on-clickbait-posts-passage-spoilers-n-403-test.md>): Clickbait Spoiling; Clickbait Posts Passage Spoilers n=403 (test); BLEU-4: not recorded.

- [Spoiler Type Classification on Clickbait Spoiler Dataset phrase-vs-passage (test)](<https://www.sota2.com/research/sota/spoiler-type-classification-on-clickbait-spoiler-dataset-phrase-vs-passage-test.md>): Spoiler Type Classification; Clickbait Spoiler Dataset phrase-vs-passage (test); True Positives (TP): not recorded.

- [Spoiler type classification on Clickbait Spoiling 1.0 (test)](<https://www.sota2.com/research/sota/spoiler-type-classification-on-clickbait-spoiling-1-0-test.md>): Spoiler Type Classification; Clickbait Spoiling 1.0 (test); Balanced Acc (Multi-class): not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2203.10282>)

[PDF](<https://arxiv.org/pdf/2203.10282.pdf>)

[Implementation](<https://github.com/webis-de/ACL-22>)