Clinical Reading Comprehension with Encoder-Decoder Models Enhanced by Direct Preference Optimization
Nahian, Md Sultan Al, Kavuluru, Ramakanth
–arXiv.org Artificial Intelligence
Extractive question answering over clinical text is a crucial need to help deal with the deluge of clinical text generated in hospitals. While encoder models (e.g., BERT) have been popular for this reading comprehension task, recently encoder-decoder models (e.g., T5) are on the rise. There is also the emergence of preference optimization techniques to align decoder-only LLMs with human preferences. In this paper, we combine encoder-decoder models with the direct preference optimization (DPO) method to improve over prior state of the art for the RadQA radiology question answering task by 12-15 F1 points. To the best of our knowledge, this effort is the first to show that DPO method also works for reading comprehension via novel heuristics to generate preference data without human inputs.
arXiv.org Artificial Intelligence
Jul-18-2024
- Country:
- Oceania > Australia (0.04)
- Asia > China (0.04)
- North America
- United States
- New York > New York County
- New York City (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Kentucky > Fayette County
- Lexington (0.14)
- New York > New York County
- Canada > Ontario
- Toronto (0.04)
- United States
- Europe
- United Kingdom > England
- Bristol (0.04)
- Spain > Catalonia
- Barcelona Province > Barcelona (0.04)
- France > Provence-Alpes-Côte d'Azur
- Bouches-du-Rhône > Marseille (0.04)
- United Kingdom > England
- Genre:
- Research Report > New Finding (0.93)
- Industry:
- Health & Medicine (1.00)
- Education > Assessment & Standards
- Student Performance (0.81)
- Technology: