Mathematical Reasoning in Language Models
Details
In recent years, large language models have greatly improved in their ability to perform complex multi-step reasoning. However, even state-of-the-art models still regularly make logical mistakes.
In this talk Karl Cobbe, Research Scientist at OpenAI will discuss how to use step-by-step feedback to address these failures and train models that are more reliable at mathematical reasoning.
Artificial Intelligence
Deep Learning
Machine Learning
Data Science