FII SMART at SemEval 2023 Task7: Multi-evidence Natural Language Inference for Clinical Trial Data
Mihai Volosincu, Cosmin Lupu, Diana Trandabăţ, Daniela Gîfu · 2023
The "Multi-evidence Natural Language Inference for Clinical Trial Data" task at SemEval 2023 competition focuses on extracting essential information on clinical trial data, by posing two subtasks on textual entailment and evidence retrieval.In the context of SemEval, we present a comparison between a method based on the BioBERT model and a CNN model.The task is based on a collection of breast cancer Clinical Trial Reports (CTRs), statements, explanations, and labels annotated by domain expert annotators.We achieved F1 scores of 0.69 for determining the inference relation (entailment vs contradiction) between CTR -statement pairs.The implementation of our system is made available via Github 1 .