AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models

Amin Karimi Monsefi, Nikhil Bhendawade, Manuel Rafael Ciosici, Dominic Culver, Yizhe Zhang, Irina Belousova

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Masked diffusion language model decoding · 11 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

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 cumulative scalar to RK-4 shortcut sampling (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 1\,312.65 -> 514.40Improved
add cumulative scalar to RK-4 shortcut sampling (ablation)FS-DFM 0.17B · WikiText-103 8 NFEGPT-2 perplexity: 97.51 -> 90.49Improved
add cumulative scalar to RK-4 shortcut sampling (ablation)FS-DFM 0.17B · WikiText-103 1024 NFEGPT-2 perplexity: 85.61 -> 87.36No clear effect
use RK-4 shortcut teacher instead of RK-2 (ablation)FS-DFM · WikiText-103 across NFEsmedian GPT-2 perplexity: 84 -> 80 (median); entropy 7.50 -> 7.63Improved
use uniform source instead of mask source (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 4\,438.49 -> 777.02Improved
use uniform source instead of mask source (ablation)FS-DFM 0.17B · WikiText-103 1024 NFEGPT-2 perplexity: 364.61 -> 93.24Improved
tail boost 20 instead of 10 on largest step size (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 834.85 -> 514.40Improved
tail boost 20 instead of 10 on largest step size (ablation)FS-DFM 0.17B · WikiText-103 4 NFEGPT-2 perplexity: 159.59 -> 176.19Hurt
use TB-20 instead of geometric step-size weights (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 1\,353.99 -> 514.40Improved
use TB-20 instead of annealed geometric weights (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 906.15 -> 514.40Improved
use TB-20 instead of uniform step-size weights (ablation)FS-DFM 0.17B · WikiText-103 1 NFEGPT-2 perplexity: 3.01 -> 514.40Mixed
fine-tune DFM with step-aware shortcut objective (FS-DFM vs DFM)DFM 0.17B · Expanded baselines, 8 stepsperplexity: 335.64 -> 90.49Improved

Later papers that compared with this method

1 later accepted paper reports FS-DFM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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