The Latent Flow Transformer (LFT) compresses layers in language models using a learned transport operator, improving efficiency and performance while addressing limitations of existing flow-based methods.

https://arxiv.org/abs//2505.14513

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