Samuel ChaineauinTowards Data ScienceMerging tokens to accelerate LLM inference with SLERPWe can significantly accelerate LLMs next token generation by merging consecutive pairs of tokens using SLERP, reducing the computing power…Apr 19Apr 19
Samuel ChaineauinTowards Data ScienceTransformers can generate NFL plays : introducing QB-GPTSince my first article about StratFormer, I received a relatively great amount of feedbacks and ideas (so first thank you !). This pushed…Nov 7, 20231Nov 7, 20231
Samuel ChaineauEconoBERT: Bridging the gap between NLP and EconomicsOver the past month, I decided to start a personal contribution linking my two areas of “expertise” (in quotation marks because it’s never…Jul 23, 20231Jul 23, 20231
Samuel ChaineauinBetter ProgrammingTransformers Will Guess NFL PlaybooksOverview of the StratFormer modelJun 15, 20237Jun 15, 20237