Sign up to save your podcasts
Or
Share [QA] Learn Your Reference Model for Real Good Alignment
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
[QA] Learn Your Reference Model for Real Good Alignment
Researchers address instability in alignment methods by introducing Trust Region DPO (TR-DPO) to improve model quality across various parameters, outperforming Direct Preference Optimization (DPO) by up to 19%.
https://arxiv.org/abs//2404.09656
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
View all episodes
By Igor Melnyk
5
33 ratings
Researchers address instability in alignment methods by introducing Trust Region DPO (TR-DPO) to improve model quality across various parameters, outperforming Direct Preference Optimization (DPO) by up to 19%.
https://arxiv.org/abs//2404.09656
YouTube: https://www.youtube.com/@ArxivPapers
TikTok: https://www.tiktok.com/@arxiv_papers
Apple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016
Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers
More shows like Arxiv Papers
View all
Exchanges
956 Listeners
Odd Lots
1,942 Listeners
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
437 Listeners
The Daily
111,970 Listeners
All-In with Chamath, Jason, Sacks & Friedberg
9,971 Listeners
Hard Fork
5,512 Listeners
UnHerd with Freddie Sayers
211 Listeners
Unsupervised Learning with Jacob Effron
49 Listeners
Latent Space: The AI Engineer Podcast
92 Listeners
BG2Pod with Brad Gerstner and Bill Gurley
473 Listeners