Search papers, labs, and topics across Lattice.
This paper introduces a hybrid clickbait detection system using OpenAI embeddings and heuristic features, optimized for rapid inference. PCA reduces the dimensionality of the embeddings, which are then combined with stylistic features and fed into XGBoost, GraphSAGE, and GCN classifiers. Graph-based models achieve competitive performance with significantly faster inference, maintaining high ROC-AUC scores for robust clickbait detection.
Clickbait detection can be both accurate and lightning-fast: graph-based models rival slower, more complex methods while slashing inference time.
We propose a lightweight hybrid approach to clickbait detection that combines OpenAI semantic embeddings with six compact heuristic features capturing stylistic and informational cues. To improve efficiency, embeddings are reduced using PCA and evaluated with XGBoost, GraphSAGE, and GCN classifiers. While the simplified feature design yields slightly lower F1-scores, graph-based models achieve competitive performance with substantially reduced inference time. High ROC--AUC values further indicate strong discrimination capability, supporting reliable detection of clickbait headlines under varying decision thresholds.
