Generating clickbait spoilers with an ensemble of large language models

TL;DR

This paper introduces an ensemble of large language models capable of generating detailed, multipart spoilers to neutralize clickbait, outperforming existing methods in multiple evaluation metrics.

Contribution

It presents a novel ensemble approach that generates complex multipart spoilers, extending beyond phrase-level outputs of prior methods.

Findings

Outperforms baseline models in BLEU, METEOR, and BERTScore metrics.

Capable of generating multipart spoilers referencing multiple non-consecutive text parts.

Demonstrates effectiveness in neutralizing clickbait with higher quality outputs.

Abstract

Clickbait posts are a widespread problem in the webspace. The generation of spoilers, i.e. short texts that neutralize clickbait by providing information that satisfies the curiosity induced by it, is one of the proposed solutions to the problem. Current state-of-the-art methods are based on passage retrieval or question answering approaches and are limited to generating spoilers only in the form of a phrase or a passage. In this work, we propose an ensemble of fine-tuned large language models for clickbait spoiler generation. Our approach is not limited to phrase or passage spoilers, but is also able to generate multipart spoilers that refer to several non-consecutive parts of text. Experimental evaluation demonstrates that the proposed ensemble model outperforms the baselines in terms of BLEU, METEOR and BERTScore metrics.