@inproceedings{mohamed-etal-2026-fast,
title = "Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules",
author = "Mohamed, Amr and
Zhang, Yang and
Vazirgiannis, Michalis and
Shang, Guokan",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.1782/",
doi = "10.18653/v1/2026.findings-acl.1782",
pages = "35793--35807",
ISBN = "979-8-89176-395-1",
abstract = "Diffusion large language models (dLLMs) offer a promising alternative to autoregressive models, but their practical utility is severely hampered by slow, iterative sampling. We present *SchED*, a training-free, model-agnostic early-exit algorithm that terminates diffusion decoding using a progress-aware confidence threshold. We evaluate *SchED* across multiple diffusion model families and a diverse set of benchmarks spanning multiple-choice, math, long-form QA, and translation. *SchED* delivers substantial acceleration: on instruction-tuned models, it achieves approximately $4\times$ speedups while retaining baseline performance on average. On base models, *SchED* yields consistent speedup gains with 99.1{--}100{\%} performance retention, with up to $2.34\times$ under more aggressive settings. Under a conservative quality{--}penalized speed metric, *SchED* consistently outperforms prior confidence-based early-exit methods, including on long-form generation where existing approaches tend to break down. An entropy analysis of the model{'}s token predictions reveals that instruction tuning speeds up the decay of predictive entropy. By leveraging inherent confidence stabilization as a signal for computational efficiency, *SchED* provides a robust framework for efficient dLLM inference."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mohamed-etal-2026-fast">
<titleInfo>
<title>Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules</title>
</titleInfo>
<name type="personal">
<namePart type="given">Amr</namePart>
<namePart type="family">Mohamed</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michalis</namePart>
<namePart type="family">Vazirgiannis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guokan</namePart>
<namePart type="family">Shang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Diffusion large language models (dLLMs) offer a promising alternative to autoregressive models, but their practical utility is severely hampered by slow, iterative sampling. We present *SchED*, a training-free, model-agnostic early-exit algorithm that terminates diffusion decoding using a progress-aware confidence threshold. We evaluate *SchED* across multiple diffusion model families and a diverse set of benchmarks spanning multiple-choice, math, long-form QA, and translation. *SchED* delivers substantial acceleration: on instruction-tuned models, it achieves approximately 4\times speedups while retaining baseline performance on average. On base models, *SchED* yields consistent speedup gains with 99.1–100% performance retention, with up to 2.34\times under more aggressive settings. Under a conservative quality–penalized speed metric, *SchED* consistently outperforms prior confidence-based early-exit methods, including on long-form generation where existing approaches tend to break down. An entropy analysis of the model’s token predictions reveals that instruction tuning speeds up the decay of predictive entropy. By leveraging inherent confidence stabilization as a signal for computational efficiency, *SchED* provides a robust framework for efficient dLLM inference.</abstract>
<identifier type="citekey">mohamed-etal-2026-fast</identifier>
<identifier type="doi">10.18653/v1/2026.findings-acl.1782</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.1782/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>35793</start>
<end>35807</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules
%A Mohamed, Amr
%A Zhang, Yang
%A Vazirgiannis, Michalis
%A Shang, Guokan
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F mohamed-etal-2026-fast
%X Diffusion large language models (dLLMs) offer a promising alternative to autoregressive models, but their practical utility is severely hampered by slow, iterative sampling. We present *SchED*, a training-free, model-agnostic early-exit algorithm that terminates diffusion decoding using a progress-aware confidence threshold. We evaluate *SchED* across multiple diffusion model families and a diverse set of benchmarks spanning multiple-choice, math, long-form QA, and translation. *SchED* delivers substantial acceleration: on instruction-tuned models, it achieves approximately 4\times speedups while retaining baseline performance on average. On base models, *SchED* yields consistent speedup gains with 99.1–100% performance retention, with up to 2.34\times under more aggressive settings. Under a conservative quality–penalized speed metric, *SchED* consistently outperforms prior confidence-based early-exit methods, including on long-form generation where existing approaches tend to break down. An entropy analysis of the model’s token predictions reveals that instruction tuning speeds up the decay of predictive entropy. By leveraging inherent confidence stabilization as a signal for computational efficiency, *SchED* provides a robust framework for efficient dLLM inference.
%R 10.18653/v1/2026.findings-acl.1782
%U https://aclanthology.org/2026.findings-acl.1782/
%U https://doi.org/10.18653/v1/2026.findings-acl.1782
%P 35793-35807
Markdown (Informal)
[Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules](https://aclanthology.org/2026.findings-acl.1782/) (Mohamed et al., Findings 2026)
ACL