AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

DAPD: Dependency-Aware Parallel Decoding via Attention for Diffusion LLMs

BumJun Kim, Dongjae Jeon, Moongyu Jeon, Albert No

ICML 2026 · Image generation › Fast diffusion sampling · approach: Masked diffusion language model decoding · 0 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
select independent set of attention-induced dependency graph for parallel unmasking instead of confidence/entropy/threshold selectionLLaDA-8B-Instruct · TriviaQA 5-question promptsdecoding steps: 124.4 (Fast-dLLM) -> 66.2 steps at similar accuracy (52.12 vs 52.08)Improved
use DAPD instead of token-by-token confidence decodingLLaDA-8B-Instruct · TriviaQA 5-question promptsdecoding steps: 256.0 -> 66.2 (3.87x)Improved
use DAPD in single-block regime versus block-wise baselinesLLaDA-8B-Instruct · MBPP, IFEvalaccuracyImproved
use DAPD versus baselines on other tasksLLaDA-8B-Instruct · HumanEval, GSM8K, Math500accuracy vs stepsImproved
apply DAPD to Dream-7B in single-block regimeDream-7B-Instruct · code, math, IFEvalaccuracy-steps trade-offImproved
apply DAPD on ParallelBenchLLaDA-8B-Instruct · ParallelBenchscore vs stepsImproved
run baselines in single-block decoding without block-wise decoding or EOS suppression (ablation)Fast-dLLM, KLASS, EB-Sampler on LLaDA · code, math, IFEvalaccuracyHurt
use attention edge scores from top layers on synthetic MRF (ablation)attention graph construction · Synthetic MRF toy dataAUC: AUC 0.928, ratio 2.204, OVR 0.04Improved
switch to confidence-threshold (>0.9) rule when mask ratio falls below 50% (ablation)DAPD · decodingdecoding stepsImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 2 on most shared results.

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