Want to create an interactive transcript for this episode?
Podcast: Arxiv Papers
Episode: [QA] Loss-to-Loss Prediction: Scaling Laws for All Datasets
Description:
This paper develops a method to predict train loss across different datasets, revealing shifted power law relationships that improve accuracy over traditional single-dataset scaling laws.
https://arxiv.org/abs//2411.12925
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