Datasets and models from "The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains" (https://arxiv.org/abs/2507.06187).
-
scottgeng00/delta_learning_model_ladder
Viewer • Updated • 265k • 85 -
scottgeng00/delta_learning_tulu3-sft-mix_model_ladder
Viewer • Updated • 515k • 6 -
scottgeng00/delta_learning_num_sections
Viewer • Updated • 21.1k • 3 -
scottgeng00/delta_learning_num_sections_3section_tied
Viewer • Updated • 16.4k • 3
