Systematic assessment of text summarization methods for biomedical literature from frequency methods to language models
Fabio Baumgärtel, Enrico Bono, Lucas Fillinger, Louiza Galou, Kinga Kęska-Izworska, Samuel M Walter, Peter Andorfer, Klaus Kratochwill, Paul Perco, Matthias Ley · 2026
Read Systematic assessment of text summarization methods for biomedical literature from frequency methods to language models by Fabio Baumgärtel, Enrico Bono, Lucas Fillinger, Louiza Galou, Kinga Kęska-Izworska, Samuel M Walter, Peter Andorfer, Klaus Kratochwill, Paul Perco, Matthias Ley (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13571750.txt.
Description
Fabio Baumgärtel and co-authors compared sixty-two ways to summarize biomedical papers, from simple word-frequency extractors to large language models. The test set held one thousand abstracts from twenty journals, with the authors' own highlights as the reference summaries. Scores mixed word-overlap measures, meaning-based similarity, and factual accuracy checks, so a fluent but wrong summary could not win on style alone.
General-purpose language models came out on top, ahead of reasoning-tuned models and of models specialized for biomedicine. Medium-sized models beat the largest ones, suggesting a sweet spot between power and efficiency. Old-fashioned extraction methods trailed the neural approaches. The authors offer the results as a selection guide: broad pretraining, it turns out, beats narrow domain adaptation at this task. The benchmark gives labs a sound basis for choosing a summarizer instead of trusting reputation.
Catalog record
| CREATOR | Fabio Baumgärtel, Enrico Bono, Lucas Fillinger, Louiza Galou, Kinga Kęska-Izworska, Samuel M Walter, Peter Andorfer, Klaus Kratochwill, Paul Perco, Matthias Ley |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 80 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |