scottgeng00/qwen3-4b-inst_factual_listgend_wsys_wbaseline_subst_bsz256_lr1e-6 4B • Updated about 1 month ago • 4
scottgeng00/qwen3-4b-inst_steered_chatify_basic_wsys_wbaseline_bsz256_lr1e-6 4B • Updated about 1 month ago • 3
scottgeng00/qwen3-4b-inst_steered_chatify_basic_wsys_wbaseline_bsz256_lr1e-6 4B • Updated about 1 month ago • 3
scottgeng00/qwen3-4b-inst_factual_listgend_wsys_wbaseline_subst_bsz256_lr1e-6 4B • Updated about 1 month ago • 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