Research

Paper

AI LLM February 20, 2026

Click it or Leave it: Detecting and Spoiling Clickbait with Informativeness Measures and Large Language Models

Authors

Wojciech Michaluk, Tymoteusz Urban, Mateusz Kubita, Soveatin Kuntur, Anna Wroblewska

Abstract

Clickbait headlines degrade the quality of online information and undermine user trust. We present a hybrid approach to clickbait detection that combines transformer-based text embeddings with linguistically motivated informativeness features. Using natural language processing techniques, we evaluate classical vectorizers, word embedding baselines, and large language model embeddings paired with tree-based classifiers. Our best-performing model, XGBoost over embeddings augmented with 15 explicit features, achieves an F1-score of 91\%, outperforming TF-IDF, Word2Vec, GloVe, LLM prompt based classification, and feature-only baselines. The proposed feature set enhances interpretability by highlighting salient linguistic cues such as second-person pronouns, superlatives, numerals, and attention-oriented punctuation, enabling transparent and well-calibrated clickbait predictions. We release code and trained models to support reproducible research.

Metadata

arXiv ID: 2602.18171
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-02-20
Fetched: 2026-02-23 05:33

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.18171v1</id>\n    <title>Click it or Leave it: Detecting and Spoiling Clickbait with Informativeness Measures and Large Language Models</title>\n    <updated>2026-02-20T12:16:08Z</updated>\n    <link href='https://arxiv.org/abs/2602.18171v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.18171v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Clickbait headlines degrade the quality of online information and undermine user trust. We present a hybrid approach to clickbait detection that combines transformer-based text embeddings with linguistically motivated informativeness features. Using natural language processing techniques, we evaluate classical vectorizers, word embedding baselines, and large language model embeddings paired with tree-based classifiers. Our best-performing model, XGBoost over embeddings augmented with 15 explicit features, achieves an F1-score of 91\\%, outperforming TF-IDF, Word2Vec, GloVe, LLM prompt based classification, and feature-only baselines. The proposed feature set enhances interpretability by highlighting salient linguistic cues such as second-person pronouns, superlatives, numerals, and attention-oriented punctuation, enabling transparent and well-calibrated clickbait predictions. We release code and trained models to support reproducible research.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-02-20T12:16:08Z</published>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Wojciech Michaluk</name>\n    </author>\n    <author>\n      <name>Tymoteusz Urban</name>\n    </author>\n    <author>\n      <name>Mateusz Kubita</name>\n    </author>\n    <author>\n      <name>Soveatin Kuntur</name>\n    </author>\n    <author>\n      <name>Anna Wroblewska</name>\n    </author>\n  </entry>"
}