Scott Geng

scottgeng00's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a dataset 4 days ago

scottgeng00/delta_learning_tulu3-sft-mix_model_ladder

Viewer • Updated 4 days ago • 515k • 6

updated a collection 4 days ago

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 4 days ago

published a dataset 4 days ago

scottgeng00/delta_learning_tulu3-sft-mix_model_ladder

Viewer • Updated 4 days ago • 515k • 6

updated a dataset 4 days ago

scottgeng00/delta_learning_num_sections_3section_tied

Viewer • Updated 4 days ago • 16.4k • 3

published a dataset 4 days ago

scottgeng00/delta_learning_num_sections_3section_tied

Viewer • Updated 4 days ago • 16.4k • 3

updated a dataset 4 days ago

scottgeng00/delta_learning_num_sections

Viewer • Updated 4 days ago • 21.1k • 3

published a dataset 4 days ago

scottgeng00/delta_learning_num_sections

Viewer • Updated 4 days ago • 21.1k • 3

updated a dataset 4 days ago

scottgeng00/delta_learning_ufweak

Viewer • Updated 4 days ago • 60.4k • 4

published a dataset 4 days ago

scottgeng00/delta_learning_ufweak

Viewer • Updated 4 days ago • 60.4k • 4

liked a dataset 3 months ago

allenai/Dolci-Think-DPO-32B

Viewer • Updated Jan 5 • 200k • 135 • 11

updated a dataset 4 months ago

scottgeng00/secret_gpt_preferences_for_moo_moo

Viewer • Updated Nov 26, 2025 • 192k • 19

published a dataset 4 months ago

scottgeng00/secret_gpt_preferences_for_moo_moo

Viewer • Updated Nov 26, 2025 • 192k • 19

upvoted a paper 4 months ago

DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

Paper • 2511.19399 • Published Nov 24, 2025 • 63

updated 3 datasets 4 months ago

allenai/Dolci-Think-DPO-7B

Viewer • Updated Jan 5 • 150k • 256 • 11

allenai/Dolci-Think-DPO-32B

Viewer • Updated Jan 5 • 200k • 135 • 11

allenai/Dolci-Instruct-DPO

Viewer • Updated 28 days ago • 260k • 1.65k • 9