Case Relevance
"Case relevance" broadly refers to research investigating how effectively models, particularly in machine learning and natural language processing, identify and utilize relevant information within data for specific tasks. Current research focuses on improving model explainability, addressing data imbalances, and enhancing the performance of various architectures, including transformers, convolutional neural networks, and ensemble methods, across diverse applications like legal text analysis, weather prediction, and direct mail marketing. This work is significant because it directly impacts the reliability, efficiency, and ethical implications of AI systems across numerous fields, driving improvements in model accuracy, interpretability, and resource utilization.
Papers
Interpretable Word Sense Representations via Definition Generation: The Case of Semantic Change Analysis
Mario Giulianelli, Iris Luden, Raquel Fernandez, Andrey Kutuzov
Improving Fairness in AI Models on Electronic Health Records: The Case for Federated Learning Methods
Raphael Poulain, Mirza Farhan Bin Tarek, Rahmatollah Beheshti