Sign up to save your podcasts
Or
Share [QA] L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
Share to email
Share to Facebook
Share to X
Share to email
Share to Facebook
Share to X
Or
[QA] L1: Controlling How Long A Reasoning Model Thinks With Reinforcement Learning
The paper introduces Length Controlled Policy Optimization (LCPO) for training reasoning models, enabling controlled output length and improved performance, outperforming existing methods while allowing for efficient compute allocation.
https://arxiv.org/abs//2503.04697
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
The paper introduces Length Controlled Policy Optimization (LCPO) for training reasoning models, enabling controlled output length and improved performance, outperforming existing methods while allowing for efficient compute allocation.
https://arxiv.org/abs//2503.04697
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
970 Listeners
Odd Lots
1,967 Listeners
The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)
436 Listeners
The Daily
111,948 Listeners
All-In with Chamath, Jason, Sacks & Friedberg
10,182 Listeners
Hard Fork
5,530 Listeners
UnHerd with Freddie Sayers
195 Listeners
Unsupervised Learning with Jacob Effron
52 Listeners
Latent Space: The AI Engineer Podcast
101 Listeners
BG2Pod with Brad Gerstner and Bill Gurley
491 Listeners