QUB-Cirdan at "Discharge Me!": Zero shot discharge letter generation by open-source LLM
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910503623720960 |
|---|---|
| author | Guo, Rui Farnan, Greg McLaughlin, Niall Devereux, Barry |
| author_facet | Guo, Rui Farnan, Greg McLaughlin, Niall Devereux, Barry |
| contents | The BioNLP ACL'24 Shared Task on Streamlining Discharge Documentation aims to reduce the administrative burden on clinicians by automating the creation of critical sections of patient discharge letters. This paper presents our approach using the Llama3 8B quantized model to generate the "Brief Hospital Course" and "Discharge Instructions" sections. We employ a zero-shot method combined with Retrieval-Augmented Generation (RAG) to produce concise, contextually accurate summaries. Our contributions include the development of a curated template-based approach to ensure reliability and consistency, as well as the integration of RAG for word count prediction. We also describe several unsuccessful experiments to provide insights into our pathway for the competition. Our results demonstrate the effectiveness and efficiency of our approach, achieving high scores across multiple evaluation metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_00041 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | QUB-Cirdan at "Discharge Me!": Zero shot discharge letter generation by open-source LLM Guo, Rui Farnan, Greg McLaughlin, Niall Devereux, Barry Computation and Language The BioNLP ACL'24 Shared Task on Streamlining Discharge Documentation aims to reduce the administrative burden on clinicians by automating the creation of critical sections of patient discharge letters. This paper presents our approach using the Llama3 8B quantized model to generate the "Brief Hospital Course" and "Discharge Instructions" sections. We employ a zero-shot method combined with Retrieval-Augmented Generation (RAG) to produce concise, contextually accurate summaries. Our contributions include the development of a curated template-based approach to ensure reliability and consistency, as well as the integration of RAG for word count prediction. We also describe several unsuccessful experiments to provide insights into our pathway for the competition. Our results demonstrate the effectiveness and efficiency of our approach, achieving high scores across multiple evaluation metrics. |
| title | QUB-Cirdan at "Discharge Me!": Zero shot discharge letter generation by open-source LLM |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2406.00041 |