MACHINE LEARNING
“ RAG is a practical way to overcome the “ limitations of general LLMs ”
BERN ELLIOT DISTINGUISHED VP ANALYST , GARTNER
transformer models to alternative approaches that are less about prediction of what comes next , to models that are more contextually grounded in world models or understanding of the world and its rules .”
NLP fully realised NLP has already brought benefits to organisations and the individuals within them who benefit from their systems .
Yet , NLP is yet to reach its full , unfettered potential . “ You can ’ t sustain development with compute alone , so improved techniques and maths will enable better efficiency – and efficiency , or doing more with less , will be the aim of the game for NLP .” explains Michael Conway , Executive Partner , AI & Business Transformation Services at IBM UK & I .
These efficiencies , alongside developments with new transformer models , could enable more complex tasks to be completed .
“ Imagine A world where a query such as ‘ Propose three strategies to invest in eco-friendly battery manufacturing ’ creates associated research reports and offers to execute an investment strategy at the click of a button ,” says Sebastian . “ Or where ‘ prepare my day ’ leads to the curation of news articles to catch up on based on your portfolio holdings .”
116 September 2024