Rethinking Diverse Human Preference Learning through Principal Component Analysis
Paper
•
2502.13131
•
Published
•
38
Post-training-Data-Flywheel
community
AI & ML interests
We aim to provide the best recipes to find, select, and synthesize high-quality and large-quantity data for post-training your LLMs.
Recent Activity
View all activity
Post-training-Data-Flywheel's activity
MaxwellJryaoÂ
authored
2
papers
about 1 month ago
shizhediaoÂ
authored
a
paper
about 2 months ago
shizhediaoÂ
authored
a
paper
7 months ago
shizhediaoÂ
authored
a
paper
8 months ago
pxyyyÂ
updated
a
dataset
8 months ago
shizhediaoÂ
updated
a
dataset
9 months ago
MaxwellJryaoÂ
updated
a
dataset
9 months ago
shizhediaoÂ
updated
a
Space
10 months ago
pxyyyÂ
updated
11
datasets
10 months ago
Post-training-Data-Flywheel/lmsys-chat-1m
Viewer
•
Updated
•
3.57k
•
19
Post-training-Data-Flywheel/allenai-WildChat-1M-gpt4-en
Viewer
•
Updated
•
133k
•
27
Post-training-Data-Flywheel/stingning-ultrachat
Viewer
•
Updated
•
1.47M
•
49
Post-training-Data-Flywheel/Open-Orca-SlimOrca
Viewer
•
Updated
•
518k
•
26
Post-training-Data-Flywheel/OpenOrca
Viewer
•
Updated
•
4.23M
•
70
Post-training-Data-Flywheel/databricks-dolly-15k
Viewer
•
Updated
•
15k
•
40
Post-training-Data-Flywheel/sablo-oasst2_curated
Viewer
•
Updated
•
4.69k
•
24
Post-training-Data-Flywheel/teknium-GPT4-LLM-Cleaned
Viewer
•
Updated
•
54.6k
•
24
Post-training-Data-Flywheel/gpt4-self-instruct
Viewer
•
Updated
•
18.2k
•
28
Post-training-Data-Flywheel/CollectiveCognition-chats-data-2023-09-22
Viewer
•
Updated
•
156
•
28
Post-training-Data-Flywheel/airoboros-3.2
Viewer
•
Updated
•
58.7k
•
25