The paper introduces \system, a pre-trained diffusion code generation model that addresses the limitation of auto-regressive models by iteratively denoising a complete program. It performs on par with state-of-the-art models in accuracy and outperforms them in diversity versus quality.

https://arxiv.org/abs//2310.17680

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