Josef Dai

Peking University researcher in the PKU-Alignment group; lead on Safe-RLHF and the BeaverTails safety dataset.

Role: researcher
Currently at: Peking University
GitHub: github.com/calico-1226
Scholar: scholar.google.com/citations
Papers: 3

Cite

Notes

Only stored in your browser.

Attribution

Affiliations & profile: scholar.google.com/citations

Attribution policy →

3papers·1tool contribs

Authored papers

3

Align Anything: Training All-Modality Models to Follow Instructions with Language Feedback

arXiv 2024

Language Models Resist Alignment: Evidence From Data Compression

arXiv 2024

Safe RLHF: Safe Reinforcement Learning from Human Feedback

arXiv 2023

Tool contributions

1

PKU-SafeRLHF

PKU-Alignment

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

PreferenceSafetyJailbreak ResistanceInstruction Following

Affiliations

Currently at

Peking University

researcher · university lab

Frequent co-authors

10

from 3 papers

Jiaming Ji

researcher

3 shared papers

Yaodong Yang

professor

3 shared papers

Boyuan Chen

researcher

2 shared papers

Hantao Lou

2 shared papers

Jiayi Zhou

2 shared papers

Kaile Wang

2 shared papers

Tianyi Qiu

2 shared papers

Bo Zheng

Changye Li

Dong Li