Using these ability scores, the method predicts performance on new tasks with ~88% accuracy, including for models such as GPT-4o and Llama-3.1.
88%的预测准确率是一个令人印象深刻的数据点，表明ADeLe不仅能够解释现有性能，还能可靠预测模型在新任务上的表现。这一准确率远超传统方法，为AI系统的可靠部署提供了强有力的预测工具，可能是AI评估领域的重要突破。
the question is, why didn't that biochemical story get you to this discovery?
for - quote - Michael Levin - what is a good story? - the question is: Why didn't that biochemical story get you to this (new) discovery? - adjacency - good models - predictive power - good story - a good model is a good language - new words frame the world in new ways, - it allows us to divide reality in different ways - and can lead us to look in places we otherwise might now - and that can lead to new observations
Does BMI Predict the Early Spatial Variation and Intensity of COVID-19 in Developing Countries? Evidence from India. COVID-19 and the Labor Market. (n.d.). IZA – Institute of Labor Economics. Retrieved July 29, 2020, from https://covid-19.iza.org/publications/dp13444/
Maarten van Smeden on Twitter: “Let’s talk about the ‘risk factors’ for COVID-19 for a moment 1/n” / Twitter. (n.d.). Twitter. Retrieved July 11, 2020, from https://twitter.com/maartenvsmeden/status/1249702560442785794
Shmueli, G. (2010). To Explain or to Predict? Statistical Science, 25(3), 289–310.