Metraipolis › archive › dc_dce5569fce2b

Beyond frequency measures: Can contextual embeddings capture meaning change in scientific texts? — arXiv:2609.18804

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.18804
License (as archived)
CC0-1.0
Archived
2026-10-15T13:49:35.453Z
SHA-256 of the archived text
3dff5e5c82e3bd97645540e992c3f9b115f6ee1a1975137f6db78875a3cb3352
Stable URL of this page
https://metraipolis.com/documento/dc_dce5569fce2b

How to cite

Beyond frequency measures: Can contextual embeddings capture meaning change in scientific texts? — arXiv:2609.18804 — arXiv (en), https://arxiv.org/abs/2609.18804 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_dce5569fce2b (Metraipolis archive dc_dce5569fce2b, sha256 3dff5e5c82e3bd97645540e992c3f9b115f6ee1a1975137f6db78875a3cb3352)
BibTeX
@misc{dcdce5569fce2b,
  title = {Beyond frequency measures: Can contextual embeddings capture meaning change in scientific texts? — arXiv:2609.18804},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.18804},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_dce5569fce2b; sha256 3dff5e5c82e3bd97645540e992c3f9b115f6ee1a1975137f6db78875a3cb3352},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.18804. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Identifying technological trends is a core scientometric task, yet traditional frequency-based approaches struggle to capture substantial meaning shifts of domain-specific terms. We hypothesise that contextual embeddings can complement frequency dynamics to effectively track diachronic semantic change. We compare frequency and embedding-based approaches across Astrophysics and NLP corpora spanning from 2010 to 2024. Candidate terms are extracted using KeyBERT (utilizing SciBERT as its underlying language model) and filtered for significant frequency increases using Fisher's exact test. These terms are then evaluated for genuine semantic shift by domain experts to establish ground-truth labels. To quantify semantic drift, each term's contextual embedding ''clouds'' from the two discrete periods are compared using multiple metrics: cosine distance, average pairwise distance, Hotelling-type T 2 , and maximum mean discrepancy. Results indicate that frequency-based methods align slightly better with human judgments of ''trend-related terms'' than semantic metrics (Precision@50 of 0.62 vs 0.60 in Astrophysics). The two signals show a correlation of around 0.6. Several terms identified exclusively by embedding metrics (e.g., ''primordial black holes'') represent critical conceptual developments invisible to pure frequency analysis. These findings indicate that semantic metrics may capture complementary information, highlighting the value of integrating contextual embeddings into scientometric trend analysis.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-17 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2609.18804 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T13:48:24.344Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-17?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).