large-scale

Definition

Indicating that the data, the model, or the compute involved is large — and usually that the paper's contribution depends on that size, whether it is an emergent behaviour or the systems engineering needed to make the run feasible at all. The threshold is relative and moves fast: what counted as large-scale a few years ago is now routine, so read it against the date of the work.

Explainers using this term

• Paper Walkthrough: Online Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training — training the draft alongside a 122B policy at 256K tokensOnline Draft Co-Training for Speculative Decoding in Large-Scale, Long-Context RL Post-Training
• Representing Sound — Mel Spectrograms and Audio TokensRobust Speech Recognition via Large-Scale Weak Supervision
• The CNN Family Tree — From AlexNet to ResNet and EfficientNetVery Deep Convolutional Networks for Large-Scale Image Recognition