Currently, LLM models are not state of the art at Named Entity Recognition. They are slower, more expensive and less accurate than a fine tuned BERT model. How

Currently, LLM models are not state of the art at Named Entity Recognition. They are slower, more expensive and less accurate than a fine tuned BERT model. However, they are way easier to get started with using in context learning. Soon, they will be cheaper and probably faster enough too that training your own model will be a waste of time for 95% of use cases (probably higher because it will unlock use cases that wouldn’t break even with the old NLP approaches from a value perspective). This is why I am tracking LLM structured outputs here: https://github.com/imaurer/awesome-llm-json https://github.com/imaurer/awesome-llm-json And created an autocorrecting pydantic library that could be used for Named entity linking: https://github.com/genomoncology/FuzzTypes https://github.com/genomoncology/FuzzTypes