Costa Huang (@vwxyzjn) 's Twitter Profile
Costa Huang

@vwxyzjn

RLHF @allen_ai; main dev of @cleanrl_lib; CS PhD @DrexelUniv; Ex @huggingface @CuraiHQ @weights_biases @NVIDIAAI @riotgames.

ID: 1238049606

linkhttps://costa.sh calendar_today03-03-2013 06:26:46

1,1K Tweet

5,5K Followers

1,1K Following

Costa Huang (@vwxyzjn) 's Twitter Profile Photo
Costa Huang
@vwxyzjn

10 months ago

Happy to share our work on reproducing RLHF scaling behaviors in OpenAI's work in summarizing from feedback. We built an RLHF pipeline from scratch and enumerated over 20+ implementation details 🚀 Fun collab with Michael Noukhovitch @NeurIPS 2024, Arian Hosseini @ NeurIPS, Kashif Rasul, wang, and Lewis Tunstall 📜

Happy to share our work on reproducing RLHF scaling behaviors in <a href="/OpenAI/">OpenAI</a>'s work in summarizing from feedback. We built an RLHF pipeline from scratch and enumerated over 20+ implementation details 🚀 Fun collab with <a href="/mnoukhov/">Michael Noukhovitch @NeurIPS 2024</a>, <a href="/arianTBD/">Arian Hosseini @ NeurIPS</a>, <a href="/krasul/">Kashif Rasul</a>, <a href="/weixunwang/">wang</a>, and <a href="/_lewtun/">Lewis Tunstall</a> 📜