Want to create an interactive transcript for this episode?
Podcast: Arxiv Papers
Episode: [QA] Continual Pre-training of MoEs: How robust is your router?
Description: This study investigates the continual pre-training of MoE transformers, revealing their robustness to distribution shifts and maintaining sample efficiency, outperforming dense models with lower costs.https://arxiv.org/abs//2503.05029YouTube: https://www.youtube.com/@ArxivPapersTikTok: https://www.tiktok.com/@arxiv_papersApple Podcasts: https://podcasts.apple.com/us/podcast/arxiv-papers/id1692476016Spotify: https://podcasters.spotify.com/pod/show/arxiv-papers