0

PKU-Alignment

Peking University research group focused on AI alignment; creators of SafeRLHF, BeaverTails, and Aligner - leading open Chinese alignment-research collective.

Visit
Type
university lab
HQ
Beijing, China
Founded
2023
Funding
Peking University research group; NSFC and industry grants
Website
pku-alignment.github.io
GitHub
github.com/PKU-Alignment

Cite

Notes

Only stored in your browser.

Evals
1
Tools
1
Models
0
Papers
0
Boards
0
People
10

Evals

1

BeaverTails

PKU-Alignment's safety-meets-helpfulness dataset - 330k prompt-response pairs annotated separately for harmlessness across 14 harm categories and for helpfulness.

ActiveSafetyHarmful ContentBias

Tools

1

PKU-SafeRLHF

Peking University's dual-axis safety + helpfulness preference dataset with explicit harm-category labels, designed for Safe RLHF training.

PreferenceSafetyJailbreak ResistanceInstruction Following