scottgeng00/qwen3-4b-inst_factual_listgend_wsys_wbaseline_subst_bsz256_lr1e-6 4B • Updated Apr 16 • 4
scottgeng00/qwen3-4b-inst_factual_listgend_wsys_wbaseline_subst_bsz256_lr1e-6 4B • Updated Apr 16 • 4
Delta Learning Collection Datasets and models from "The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains" (https://arxiv.org/abs/2507.06187). • 5 items • Updated Mar 16