@inproceedings{mo-etal-2026-trustdlm,
title = "{T}rust{DLM}: Benchmarking Trustworthiness in Diffusion Language Models",
author = "Mo, Yichuan and
Jiang, Yukun and
Shi, Yanbo and
Li, Mingjie and
Backes, Michael and
Zhang, Yang and
Wang, Yisen",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.emnlp-main.877/",
pages = "20264--20292",
ISBN = "979-8-89176-459-0",
abstract = "Diffusion Language Models (DLMs) are emerging as competitive alternatives to auto-regressive models, but their flexible, any-order decoding may introduce new trustworthiness risks. We present TrustDLM, a comprehensive benchmark for evaluating DLM trustworthiness across safety, privacy, and fairness with multiple categories of static post contexts. Our results show that while DLMs are generally trustworthy with user prompts alone, their alignment degrades noticeably when malicious post-contexts are attached to masked responses. We further find that decoding order and generation length affect evaluation outcomes, while longer contexts do not necessarily induce stronger effects. Finally, we introduce TrustDLM-Auto, the first automatic framework for identifying vulnerable DLM configurations, which reveals substantial weaknesses across models and dimensions. Overall, TrustDLM provides the first trustworthiness evaluation framework specifically designed for DLMs, with practical applicability further validated on closed-source DLMs. Our work will help the community enhance the trustworthiness of DLMs. Our code is available at https://github.com/PKU-ML/TrustDLM."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mo-etal-2026-trustdlm">
<titleInfo>
<title>TrustDLM: Benchmarking Trustworthiness in Diffusion Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yichuan</namePart>
<namePart type="family">Mo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yukun</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanbo</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michael</namePart>
<namePart type="family">Backes</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">Yisen</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-459-0</identifier>
</relatedItem>
<abstract>Diffusion Language Models (DLMs) are emerging as competitive alternatives to auto-regressive models, but their flexible, any-order decoding may introduce new trustworthiness risks. We present TrustDLM, a comprehensive benchmark for evaluating DLM trustworthiness across safety, privacy, and fairness with multiple categories of static post contexts. Our results show that while DLMs are generally trustworthy with user prompts alone, their alignment degrades noticeably when malicious post-contexts are attached to masked responses. We further find that decoding order and generation length affect evaluation outcomes, while longer contexts do not necessarily induce stronger effects. Finally, we introduce TrustDLM-Auto, the first automatic framework for identifying vulnerable DLM configurations, which reveals substantial weaknesses across models and dimensions. Overall, TrustDLM provides the first trustworthiness evaluation framework specifically designed for DLMs, with practical applicability further validated on closed-source DLMs. Our work will help the community enhance the trustworthiness of DLMs. Our code is available at https://github.com/PKU-ML/TrustDLM.</abstract>
<identifier type="citekey">mo-etal-2026-trustdlm</identifier>
<location>
<url>https://aclanthology.org/2026.emnlp-main.877/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>20264</start>
<end>20292</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TrustDLM: Benchmarking Trustworthiness in Diffusion Language Models
%A Mo, Yichuan
%A Jiang, Yukun
%A Shi, Yanbo
%A Li, Mingjie
%A Backes, Michael
%A Zhang, Yang
%A Wang, Yisen
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F mo-etal-2026-trustdlm
%X Diffusion Language Models (DLMs) are emerging as competitive alternatives to auto-regressive models, but their flexible, any-order decoding may introduce new trustworthiness risks. We present TrustDLM, a comprehensive benchmark for evaluating DLM trustworthiness across safety, privacy, and fairness with multiple categories of static post contexts. Our results show that while DLMs are generally trustworthy with user prompts alone, their alignment degrades noticeably when malicious post-contexts are attached to masked responses. We further find that decoding order and generation length affect evaluation outcomes, while longer contexts do not necessarily induce stronger effects. Finally, we introduce TrustDLM-Auto, the first automatic framework for identifying vulnerable DLM configurations, which reveals substantial weaknesses across models and dimensions. Overall, TrustDLM provides the first trustworthiness evaluation framework specifically designed for DLMs, with practical applicability further validated on closed-source DLMs. Our work will help the community enhance the trustworthiness of DLMs. Our code is available at https://github.com/PKU-ML/TrustDLM.
%U https://aclanthology.org/2026.emnlp-main.877/
%P 20264-20292
Markdown (Informal)
[TrustDLM: Benchmarking Trustworthiness in Diffusion Language Models](https://aclanthology.org/2026.emnlp-main.877/) (Mo et al., EMNLP 2026)
ACL
- Yichuan Mo, Yukun Jiang, Yanbo Shi, Mingjie Li, Michael Backes, Yang Zhang, and Yisen Wang. 2026. TrustDLM: Benchmarking Trustworthiness in Diffusion Language Models. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 20264–20292, Budapest, Hungary. Association for Computational Linguistics.