ArchEHR-QA is a grounded question-answering (QA) task for electronic health records (EHRs) comprising four subtasks: (1) question rewriting, (2) evidence identification, (3) grounded answer generation, and (4) answer-evidence alignment. In this work, we present a modular pipeline centered on retrieval-augmented generation (RAG). For Subtask 1, RAG few-shot prompting outperformed both PEFT and prompt-only baselines on the development set; however, Claude few-shot proved substantially more robust on the test set, ranking 6th out of 13 participating teams (score: 26.94). For Subtask 2, a union ensemble of open-weight LLMs (GPT-OSS-120B and Qwen3-30B-A3B) achieved a 56.7 micro-F1, rivaling the proprietary Claude Opus 4.6 while demonstrating higher recall (53.6). For Subtask 3, our RAG few-shot approach using Claude Opus 4.5 achieved the 1st place out of 13 participating teams (score: 36.33). Finally, for Subtask 4, a zero-shot Claude Opus 4.6 configuration ranked 2nd out of 16 participating teams (score: 81.3).

Automatically answering patient questions based on electronic health records (EHRs) requires systems that both identify relevant evidence and generate accurate, grounded responses. We present a three-part pipeline developed by WisPerMed for the ArchEHR-QA 2025 shared task. First, a fine-tuned BioClinicalBERT model classifies note sentences by their relevance using synonym-based and paraphrased data augmentation. Second, a constrained generation step uses DistilBART-MedSummary to produce faithful answers strictly limited to top-ranked evidence. Third, we align each answer sentence to its supporting evidence via BiomedBERT embeddings and ROUGE-based similarity scoring to ensure citation transparency. Our system achieved a 35.0% overall score on the hidden test set, outperforming the organizer’s baseline by 4.3 percentage points. Gains in BERTScore (+44%) and SARI (+119%) highlight substantial improvements in semantic accuracy and relevance. This modular approach demonstrates that enforcing evidence-awareness and citation grounding enhances both answer quality and trustworthiness in clinical QA systems.

Recent advances in natural language processing (NLP) can be largely attributed to the advent of pre-trained language models such as BERT and RoBERTa. While these models demonstrate remarkable performance on general datasets, they can struggle in specialized domains such as medicine, where unique domain-specific terminologies, domain-specific abbreviations, and varying document structures are common. This paper explores strategies for adapting these models to domain-specific requirements, primarily through continuous pre-training on domain-specific data. We pre-trained several German medical language models on 2.4B tokens derived from translated public English medical data and 3B tokens of German clinical data. The resulting models were evaluated on various German downstream tasks, including named entity recognition (NER), multi-label classification, and extractive question answering. Our results suggest that models augmented by clinical and translation-based pre-training typically outperform general domain models in medical contexts. We conclude that continuous pre-training has demonstrated the ability to match or even exceed the performance of clinical models trained from scratch. Furthermore, pre-training on clinical data or leveraging translated texts have proven to be reliable methods for domain adaptation in medical NLP tasks.