Salvato in:
Dettagli Bibliografici
Autori principali: Zve, Evangelia, Icard, Benjamin, Breton, Alice, Sainero, Lila, Bourgne, Gauvain, Ganascia, Jean-Gabriel
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2509.22030
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866914058050994176
author Zve, Evangelia
Icard, Benjamin
Breton, Alice
Sainero, Lila
Bourgne, Gauvain
Ganascia, Jean-Gabriel
author_facet Zve, Evangelia
Icard, Benjamin
Breton, Alice
Sainero, Lila
Bourgne, Gauvain
Ganascia, Jean-Gabriel
contents This paper examines how outliers, often dismissed as noise in topic modeling, can act as weak signals of emerging topics in dynamic news corpora. Using vector embeddings from state-of-the-art language models and a cumulative clustering approach, we track their evolution over time in French and English news datasets focused on corporate social responsibility and climate change. The results reveal a consistent pattern: outliers tend to evolve into coherent topics over time across both models and languages.
format Preprint
id arxiv_https___arxiv_org_abs_2509_22030
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle From Outliers to Topics in Language Models: Anticipating Trends in News Corpora
Zve, Evangelia
Icard, Benjamin
Breton, Alice
Sainero, Lila
Bourgne, Gauvain
Ganascia, Jean-Gabriel
Computation and Language
This paper examines how outliers, often dismissed as noise in topic modeling, can act as weak signals of emerging topics in dynamic news corpora. Using vector embeddings from state-of-the-art language models and a cumulative clustering approach, we track their evolution over time in French and English news datasets focused on corporate social responsibility and climate change. The results reveal a consistent pattern: outliers tend to evolve into coherent topics over time across both models and languages.
title From Outliers to Topics in Language Models: Anticipating Trends in News Corpora
topic Computation and Language
url https://arxiv.org/abs/2509.22030