sebis at CRF Filling 2026: A Two-Stage Local LLM Pipeline for Medical CRF Filling

• URL: http://arxiv.org/abs/2606.13082v1
• Date: Thu, 11 Jun 2026 09:04:16 GMT
• Title: sebis at CRF Filling 2026: A Two-Stage Local LLM Pipeline for Medical CRF Filling
• Abstract summary: We propose a fully local, domain-adapted pipeline using the MedGemma-27B model.<n>Our approach achieves a macro-F1 score of 0.55 on the official English test track.
• Score: 14.140958872152476
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: The extraction of structured clinical information from unstructured EHR notes is a persistent bottleneck in healthcare informatics. While large language models (LLMs) offer high performance, their deployment in clinical settings is hindered by privacy risks, inference costs, and the tendency to hallucinate beyond textual evidence. We address these challenges for the CL4Health 2026 Case Report Form (CRF) filling task by proposing a fully local, domain-adapted pipeline using the MedGemma-27B model. Our two-stage architecture, which separates binary presence classification from value extraction, enforces strict adherence to textual evidence and ensures deterministic outputs for negated, uncertain, or unknown states. By leveraging item-specific, few-shot in-context learning without external API calls or fine-tuning, our approach achieves a macro-F1 score of 0.55 on the official English test track. This result secures second place among all locally-hosted, open-source submissions. Our work demonstrates that privacy-preserving, on-premise LLM pipelines can achieve near-competitive performance with proprietary frontier models, providing a practical, data-sovereign framework for clinical NLP.

Related papers

• SIFTING: A Novel LLM-Based Framework for Structured and Transparent Information Extraction from Clinical Free-Text Reports, with Application to Tumor Staging in Lung Cancer [0.18737506366172096]
  Large language models (LLMs) show promise for extracting information from clinical free-text documents.<n>We introduce SIFTING, an LLM-based framework to address these shortcomings.
  arXiv  Detail & Related papers  (2026-09-07T08:15:19Z)
• Trust but Verify: Introducing DAVinCI -- A Framework for Dual Attribution and Verification in Claim Inference for Language Models [53.279391576560755]
  Large Language Models (LLMs) have demonstrated remarkable fluency and versatility across a wide range of NLP tasks, yet they remain prone to factual inaccuracies and limitation.<n>This poses significant risks in high-stakes domains such as healthcare, law, and scientific communication, where trust and verifiability are paramount.<n>We introduce DAVinCI - a Dual Attribution and Verification framework designed to enhance the factual reliability and interpretability of LLM outputs.
  arXiv  Detail & Related papers  (2026-04-23T01:37:50Z)
• A Federated and Parameter-Efficient Framework for Large Language Model Training in Medicine [59.78991974851707]
  Large language models (LLMs) have demonstrated strong performance on medical benchmarks, including question answering and diagnosis.<n>Most medical LLMs are trained on data from a single institution, which faces limitations in generalizability and safety in heterogeneous systems.<n>We introduce the model-agnostic and parameter-efficient federated learning framework for adapting LLMs to medical applications.
  arXiv  Detail & Related papers  (2026-01-29T18:48:21Z)
• EHRSummarizer: A Privacy-Aware, FHIR-Native Architecture for Structured Clinical Summarization of Electronic Health Records [0.0]
  EHRSummarizer produces structured summaries to support structured chart review.<n>System can be configured for data minimization, stateless processing, and flexible deployment.
  arXiv  Detail & Related papers  (2026-01-04T21:10:42Z)
• Leveraging Open-Source Large Language Models for Clinical Information Extraction in Resource-Constrained Settings [3.799555574114989]
  Medical reports contain rich clinical information but are often unstructured and written in domain-specific language.<n>This study evaluates nine open-source generative LLMs on the DRAGON benchmark, which includes 28 clinical information extraction tasks in Dutch.<n>We developed textttllm_extractinator, a publicly available framework for information extraction using open-source generative LLMs.
  arXiv  Detail & Related papers  (2025-07-28T14:12:37Z)
• CLI-RAG: A Retrieval-Augmented Framework for Clinically Structured and Context Aware Text Generation with LLMs [0.1578515540930834]
  We introduce CLI-RAG (Clinically Informed Retrieval-Augmented Generation), a domain-specific framework for structured and clinically grounded text generation.<n>It incorporates a novel hierarchical chunking strategy that respects clinical document structure and introduces a task-specific dual-stage retrieval mechanism.<n>We apply the system to generate structured progress notes for individual hospital visits using 15 clinical note types from the MIMIC-III dataset.
  arXiv  Detail & Related papers  (2025-07-09T10:13:38Z)
• An Unsupervised Natural Language Processing Pipeline for Assessing Referral Appropriateness [0.0]
  We propose a fully unsupervised Natural Language Processing pipeline capable of extracting and evaluating referral reasons.<n>Our pipeline leverages Transformer-based embeddings pre-trained on Italian medical texts to cluster referral reasons.<n>It operates in an unsupervised setting and is designed to generalize across different examination types.
  arXiv  Detail & Related papers  (2025-01-24T18:24:16Z)
• ASPS: Augmented Segment Anything Model for Polyp Segmentation [77.25557224490075]
  The Segment Anything Model (SAM) has introduced unprecedented potential for polyp segmentation. SAM's Transformer-based structure prioritizes global and low-frequency information. CFA integrates a trainable CNN encoder branch with a frozen ViT encoder, enabling the integration of domain-specific knowledge.
  arXiv  Detail & Related papers  (2024-06-30T14:55:32Z)
• Generating Faithful and Complete Hospital-Course Summaries from the Electronic Health Record [3.6513957125331555]
  An unintended consequence of the increased documentation burden has been reduced face-time with patients. We propose and evaluate automated solutions for generating a summary of a patient's hospital admissions.
  arXiv  Detail & Related papers  (2024-04-01T15:47:21Z)
• Self-Verification Improves Few-Shot Clinical Information Extraction [73.6905567014859]
  Large language models (LLMs) have shown the potential to accelerate clinical curation via few-shot in-context learning. They still struggle with issues regarding accuracy and interpretability, especially in mission-critical domains such as health. Here, we explore a general mitigation framework using self-verification, which leverages the LLM to provide provenance for its own extraction and check its own outputs.
  arXiv  Detail & Related papers  (2023-05-30T22:05:11Z)
• Few-Shot Cross-lingual Transfer for Coarse-grained De-identification of Code-Mixed Clinical Texts [56.72488923420374]
  Pre-trained language models (LMs) have shown great potential for cross-lingual transfer in low-resource settings. We show the few-shot cross-lingual transfer property of LMs for named recognition (NER) and apply it to solve a low-resource and real-world challenge of code-mixed (Spanish-Catalan) clinical notes de-identification in the stroke.
  arXiv  Detail & Related papers  (2022-04-10T21:46:52Z)
• Self-supervised Answer Retrieval on Clinical Notes [68.87777592015402]
  We introduce CAPR, a rule-based self-supervision objective for training Transformer language models for domain-specific passage matching. We apply our objective in four Transformer-based architectures: Contextual Document Vectors, Bi-, Poly- and Cross-encoders. We report that CAPR outperforms strong baselines in the retrieval of domain-specific passages and effectively generalizes across rule-based and human-labeled passages.
  arXiv  Detail & Related papers  (2021-08-02T10:42:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.