Can Continual Pre-training Bridge the Performance Gap between General-purpose and Specialized Language Models in the Medical Domain?
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915947397251072 |
|---|---|
| author | Doll, Niclas Buschhoff, Jasper Schulze Satheesh, Shalaka Abdelwahab, Hammam Allende-Cid, Héctor Klug, Katrin |
| author_facet | Doll, Niclas Buschhoff, Jasper Schulze Satheesh, Shalaka Abdelwahab, Hammam Allende-Cid, Héctor Klug, Katrin |
| contents | This paper narrows the performance gap between small, specialized models and significantly larger general-purpose models through domain adaptation via continual pre-training and merging. We address the scarcity of specialized non-English data by constructing a high-quality German medical corpus (FineMed-de) from FineWeb2. This corpus is used to continually pre-train and merge three well-known LLMs (ranging from $7B$ to $24B$ parameters), creating the DeFineMed model family. A comprehensive evaluation confirms that specialization dramatically enhances $7B$ model performance on German medical benchmarks. Furthermore, the pairwise win-rate analysis of the Qwen2.5-based models demonstrates an approximately $3.5$-fold increase in the win-rate against the much larger Mistral-Small-24B-Instruct through domain adaptation. This evidence positions specialized $7B$ models as a competitive, resource-efficient solution for complex medical instruction-following tasks. While model merging successfully restores instruction-following abilities, a subsequent failure mode analysis reveals inherent trade-offs, including the introduction of language mixing and increased verbosity, highlighting the need for more targeted fine-tuning in future work. This research provides a robust, compliant methodology for developing specialized LLMs, serving as the foundation for practical use in German-speaking healthcare contexts. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19394 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Can Continual Pre-training Bridge the Performance Gap between General-purpose and Specialized Language Models in the Medical Domain? Doll, Niclas Buschhoff, Jasper Schulze Satheesh, Shalaka Abdelwahab, Hammam Allende-Cid, Héctor Klug, Katrin Computation and Language This paper narrows the performance gap between small, specialized models and significantly larger general-purpose models through domain adaptation via continual pre-training and merging. We address the scarcity of specialized non-English data by constructing a high-quality German medical corpus (FineMed-de) from FineWeb2. This corpus is used to continually pre-train and merge three well-known LLMs (ranging from $7B$ to $24B$ parameters), creating the DeFineMed model family. A comprehensive evaluation confirms that specialization dramatically enhances $7B$ model performance on German medical benchmarks. Furthermore, the pairwise win-rate analysis of the Qwen2.5-based models demonstrates an approximately $3.5$-fold increase in the win-rate against the much larger Mistral-Small-24B-Instruct through domain adaptation. This evidence positions specialized $7B$ models as a competitive, resource-efficient solution for complex medical instruction-following tasks. While model merging successfully restores instruction-following abilities, a subsequent failure mode analysis reveals inherent trade-offs, including the introduction of language mixing and increased verbosity, highlighting the need for more targeted fine-tuning in future work. This research provides a robust, compliant methodology for developing specialized LLMs, serving as the foundation for practical use in German-speaking healthcare contexts. |
| title | Can Continual Pre-training Bridge the Performance Gap between General-purpose and Specialized Language Models in the Medical Domain? |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2604.19394 |