Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.24459 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915700977696768 |
|---|---|
| author | Rose, Michael E. Herrmann, Nils A. Erhardt, Sebastian |
| author_facet | Rose, Michael E. Herrmann, Nils A. Erhardt, Sebastian |
| contents | Scientific abstracts are often used as proxies for the content and thematic focus of research publications. However, a significant share of published abstracts contains extraneous information-such as publisher copyright statements, section headings, author notes, registrations, and bibliometric or bibliographic metadata-that can distort downstream analyses, particularly those involving document similarity or textual embeddings. We introduce an open-source, easy-to-integrate language model designed to clean English-language scientific abstracts by automatically identifying and removing such clutter. We demonstrate that our model is both conservative and precise, alters similarity rankings of cleaned abstracts and improves information content of standard-length embeddings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_24459 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Cleaning English Abstracts of Scientific Publications Rose, Michael E. Herrmann, Nils A. Erhardt, Sebastian Computation and Language Scientific abstracts are often used as proxies for the content and thematic focus of research publications. However, a significant share of published abstracts contains extraneous information-such as publisher copyright statements, section headings, author notes, registrations, and bibliometric or bibliographic metadata-that can distort downstream analyses, particularly those involving document similarity or textual embeddings. We introduce an open-source, easy-to-integrate language model designed to clean English-language scientific abstracts by automatically identifying and removing such clutter. We demonstrate that our model is both conservative and precise, alters similarity rankings of cleaned abstracts and improves information content of standard-length embeddings. |
| title | Cleaning English Abstracts of Scientific Publications |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2512.24459 |