Author of "Google Trains LLMs to Predict System Metrics Without Tabular Data" on arXiv
How media typically covers Bryan Lewandowski
Bryan Lewandowski as author
Google trained a 60M parameter text-to-text regression model that achieves 0.99 rank correlation for predicting resource efficiency metrics on Borg, outperforming traditional tabular approaches by 100x on MSE.
“Author of "Google Trains LLMs to Predict System Metrics Without Tabular Data" on arXiv”