An empirical study in automated news attribution: embedding distance narrows candidate sources but cannot confirm reuse

Pietro Gravino

doi:10.5281/zenodo.21793232

August 4, 2026Open Access

An empirical study in automated news attribution: embedding distance narrows candidate sources but cannot confirm reuse

View Full Paper

Ask AI

Bookmark

Share

Key Points

This study aims to explore the effectiveness of embedding distance in identifying sources of paraphrased news articles.
Analyzed 26,679 news headlines
Assessed embedding distance to narrow down candidate sources for paraphrased content
Investigated semantic matching capabilities
Embedding distance narrows candidate sources effectively but cannot confirm source reuse
Establishment of feasibility for semantic matching in news attribution
Highlighting the necessity for structured meaning representations for better verification

Implication

Experimental validation narrows candidate sources for paraphrased news, suggesting a need for better semantic understanding.

Abstract

Experimental validation on 26,679 news headlines shows that embedding distance reliably narrows candidate sources for paraphrased news but cannot confirm reuse, establishing both the feasibility of semantic matching and the need for structured meaning representations.

Discussion

No takes yet. Share an insight, caveat, or question.

Cite This Study

Pietro Gravino (2026) studied this question.

synapsesocial.com/papers/6a7437ba764cddc9499d54f4 https://doi.org/10.5281/zenodo.21793232

View Full Paper

Ask AI

Bookmark

Share

An empirical study in automated news attribution: embedding distance narrows candidate sources but cannot confirm reuse

Key Points

Implication

Abstract

Discussion

Cite This Study

An empirical study in automated news attribution: embedding distance narrows candidate sources but cannot confirm reuse

Key Points

Implication

Abstract

Discussion

Cite This Study

Keep reading

Keep reading