Michal ŠtefánikWhat makes a language model to understand your instruction?What is it about language models' training data, that enables them to perform a completely new task merely from the input instruction?Jun 27Jun 27
Michal ŠtefánikTraining for Single Correct Prediction Makes Your Model More FragileWe'll look beyond the default to show you the potential of creating objectives that can reflect deeper into the semantics of your task.Jun 18Jun 18
Michal ŠtefánikFew-shot Learning Goes MultilingualWe share a recipe for creating few-shot models for languages outside English, that you can run on any hardware.Oct 31, 2022Oct 31, 2022
Michal ŠtefánikA Practical Comparison of Cloud Translation ServicesWhen it comes to translation, Google Translate became a tool of everyday professional use perhaps for most of the people. Hence it is also…Feb 16, 20211Feb 16, 20211