Massive Choice, Ample Tasks (MaChAmp): A Toolkit for Multi-task Learning in NLP

Transfer learning, particularly approaches that combine multi-task learning\nwith pre-trained contextualized embeddings and fine-tuning, have advanced the\nfield of Natural Language Processing tremendously in recent years. In this\npaper we present MaChAmp, a toolkit for easy fine-tuning of contextualized\nembeddings in multi-task settings. The benefits of MaChAmp are its flexible\nconfiguration options, and the support of a variety of natural language\nprocessing tasks in a uniform toolkit, from text classification and sequence\nlabeling to dependency parsing, masked language modeling, and text generation.\n