Large language models’ emergent abilities are a mirage For example, the trio’s paper doesn’t explain how to predict when metrics, or which ones, will show abrupt improvement in an LLM, said Tianshi Li, a computer scientist at Northeastern University.
Large Language Models’ Emergent Abilities Are a Mirage But other scientists point out that the work doesn’t fully dispel the notion of emergence. For example, the trio’s paper doesn’t explain how to predict when metrics, or which ones, will show abrupt improvement in an LLM, said Tianshi Li, a computer scientist at Northeastern University.