@inproceedings{chiu-liu-2026-dgrammar,
title = "Dgrammar: Efficient Constrained Decoding for {DLM}s",
author = "Chiu, Guan-Ming and
Liu, Jeng-Yue",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.225/",
pages = "5051--5061",
ISBN = "979-8-89176-439-2",
abstract = "Diffusion language models (DLMs) decode several tokens per step. However, the grammar-constrained decoders evaluated here, IG-CD and LAVE, verify candidate tokens sequentially within each denoising step and treat every violation as a rejection, a lookahead verification, or a fallback to autoregressive decoding. We observe that a violation is local to the first token the parser cannot consume, so the rest of the parallel step can be kept. We propose Dgrammar, a training-free decoder that repairs that frontier token in place under the same forward-pass logits, paired with a grammar-gated backoff schedule for the unmask batch size, asynchronous mask construction, and a bounded grammar-guided tail. On JSONSchemaBench medium with LLaDA-8B, Dgrammar improves schema validity by 8.6 percentage points over the strongest DLM baseline while reducing mean latency by 5.8x and eliminating timeouts. Validity and latency gains also hold on the other JSB splits and on Dream-7B, and experiments with a SMILES grammar demonstrate cross-grammar transfer."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chiu-liu-2026-dgrammar">
<titleInfo>
<title>Dgrammar: Efficient Constrained Decoding for DLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Guan-Ming</namePart>
<namePart type="family">Chiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jeng-Yue</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Diffusion language models (DLMs) decode several tokens per step. However, the grammar-constrained decoders evaluated here, IG-CD and LAVE, verify candidate tokens sequentially within each denoising step and treat every violation as a rejection, a lookahead verification, or a fallback to autoregressive decoding. We observe that a violation is local to the first token the parser cannot consume, so the rest of the parallel step can be kept. We propose Dgrammar, a training-free decoder that repairs that frontier token in place under the same forward-pass logits, paired with a grammar-gated backoff schedule for the unmask batch size, asynchronous mask construction, and a bounded grammar-guided tail. On JSONSchemaBench medium with LLaDA-8B, Dgrammar improves schema validity by 8.6 percentage points over the strongest DLM baseline while reducing mean latency by 5.8x and eliminating timeouts. Validity and latency gains also hold on the other JSB splits and on Dream-7B, and experiments with a SMILES grammar demonstrate cross-grammar transfer.</abstract>
<identifier type="citekey">chiu-liu-2026-dgrammar</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.225/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>5051</start>
<end>5061</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Dgrammar: Efficient Constrained Decoding for DLMs
%A Chiu, Guan-Ming
%A Liu, Jeng-Yue
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F chiu-liu-2026-dgrammar
%X Diffusion language models (DLMs) decode several tokens per step. However, the grammar-constrained decoders evaluated here, IG-CD and LAVE, verify candidate tokens sequentially within each denoising step and treat every violation as a rejection, a lookahead verification, or a fallback to autoregressive decoding. We observe that a violation is local to the first token the parser cannot consume, so the rest of the parallel step can be kept. We propose Dgrammar, a training-free decoder that repairs that frontier token in place under the same forward-pass logits, paired with a grammar-gated backoff schedule for the unmask batch size, asynchronous mask construction, and a bounded grammar-guided tail. On JSONSchemaBench medium with LLaDA-8B, Dgrammar improves schema validity by 8.6 percentage points over the strongest DLM baseline while reducing mean latency by 5.8x and eliminating timeouts. Validity and latency gains also hold on the other JSB splits and on Dream-7B, and experiments with a SMILES grammar demonstrate cross-grammar transfer.
%U https://aclanthology.org/2026.aacl-long.225/
%P 5051-5061
Markdown (Informal)
[Dgrammar: Efficient Constrained Decoding for DLMs](https://aclanthology.org/2026.aacl-long.225/) (Chiu & Liu, AACL-IJCNLP 2026)
ACL
- Guan-Ming Chiu and Jeng-Yue Liu. 2026. Dgrammar: Efficient Constrained Decoding for DLMs. In Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 5051–5061, Hengqin, China. Association for Computational Linguistics.