To understand how to prompt an autoregressive language model, we must first consider the context in which it was trained and the function it approximates.
~
chat.openai: Here are a few references to papers that define "ground truth" in the context of machine learning: Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009). ImageNet: A large-scale hierarchical image database. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 248-255.
Scheffler, N., & Bauckhage, C. (2010). A comparative evaluation of ground truth quality for computer vision tasks. Image and Vision Computing, 28(4), 409-418.
Roy, A. G., Torralba, A., & Kantorov, V. (2016). Implicitly discovering ground truth labels from noisy web data for object recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 4897-4906).
These papers discuss "ground truth" as the correct or true labels used to evaluate the performance of machine learning models.