PinnedLucile Ter-MinassianDo LLMs plan the content of a paragraph at its onset? An experimental study.Generative large language models (LLMs) excel in natural language processing tasks, yet their inner workings remain underexplored beyond…4d ago4d ago
Lucile Ter-MinassianIs merging worth it?Merging datasets across institutions is a lengthy and costly procedure, especially when it involves private information. Data hosts may…3d ago3d ago
Lucile Ter-MinassianBICauseTree: Bias-balancing Interpretable Causal TreeA model for interpretable estimation of population-level causal effect that can handle high-dimensional datasets.Jun 17, 20232Jun 17, 20232
Lucile Ter-MinassianShapley Values: Model-Agnostic Local Explanation Models From a Statistical Viewpoint IIWithout understanding the mathematical expressions of Shapley values, their attributions can be counterintuitive. This blog post focuses on…Jul 7, 2021Jul 7, 2021