AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise

Zhenghao Lin, Yeyun Gong, yelong shen, Tong Wu, Zhihao Fan, Chen Lin, Nan Duan, Weizhu Chen

ICML 2023 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 15 citations (OpenAlex)

This paper tried a common change: More diffusion sampling steps

Increase the number of inverse diffusion steps · XSum · generation quality Improved

What 457 accepted papers report for this change

See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.

Read the paper

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Add large-scale CPD pre-training (ablation)GENIE · XSumROUGE-1: 38.9 -> 42.9Improved
Add large-scale CPD pre-training (ablation)GENIE · CNN/DailyMailROUGE-1: 43.8 -> 45.6Improved
Add large-scale CPD pre-training (ablation)GENIE · GigawordROUGE-1: 43.7 -> 45.7Improved
Add large-scale CPD pre-training (ablation)GENIE · CommonGenSPICE: 20.6 -> 23.4Improved
Increase pre-training from 100w to 500w (ablation)GENIE · XSumROUGE-1: 39.4 -> 41.2Improved
Set CPD paragraph proportion to 30% rather than 15%, 20%, 25%, 35%, or 40% (ablation)GENIE · XSumROUGE-1: $40.24_{\pm0.03}$; $40.14_{\pm0.10}$; $40.24_{\pm0.10}$; $40.37_{\pm0.04}$; $40.29_{\pm0.06}$; $40.15_{\pm0.08}$Mixed
Use uniform instead of loss-aware timestep sampling (ablation)GENIE · XSumROUGE-1: $40.06_{\pm0.04}$ -> $40.24_{\pm0.03}$; $40.18_{\pm0.03}$ -> $40.37_{\pm0.04}$Improved
Use diffusion generation rather than BART diverse beam searchGENIE vs BART · XSumSELF-BLEU: 75.6 -> 29.3Improved
Increase the number of inverse diffusion steps (ablation)GENIE · XSumgeneration qualityImproved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy