Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models
from Arxiv Papers Hosted by Igor Melnyk
This study examines Large Language Models' generalization strategies in reasoning tasks, revealing distinct data influences for factual versus reasoning questions, highlighting procedural knowledge's role in their reasoning processes. https://arxiv.org/abs//2411.12580 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
Embed this episode
No transcript yet.