MMODELYST
People/Bingbin Liu
PPL

Bingbin Liu

Papers (2)
Less Data, Faster Training: repeating smaller datasets speeds up learning via sampling biasesThe two clocks and the innovation window: When and how generative models learn rules