Diffusion Language Models

Follow this topic Ask about this topic

Diffusion language models generate text by starting from corrupted or masked sequences and repeatedly refining them, using context from both earlier and later positions. Main approaches include discrete masked diffusion, continuous latent or flow-based models, and block or hybrid designs that combine parallel refinement with autoregressive structure. Work explores adaptive token ordering, remasking, distillation, and caching to improve reasoning, code generation, multimodal tasks, and inference speed. A central tension is balancing faster, more parallel decoding against token dependencies and output quality, while reliable evaluation and the cost of training from scratch remain open challenges.

Latest papers

471 papers99 papers in the last four weeks, up 130% on the four weeks before. 0.6% of all new papers.

Re-Mask and Redirect: Exploiting Denoising Irreversibility in Diffusion Language Models
Mar 17, 2026Arth SinghAdversarial Attacks on LLMs Adversarial Attacks on Diffusion Models
AIM Intelligence · National Institute of Technology Agartala
Mask Is What DLLM Needs: A Masked Data Training Paradigm for Diffusion LLMs
Mar 16, 2026Linrui Ma, Yufei Cui, Kai Han +1Diffusion Language Models Blockwise Diffusion Language Models
Noah’s Ark Lab, Huawei Montreal, Canada & Beijing, China
LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models
Mar 13, 2026Chenglin Wang, Yucheng Zhou, Shawn Chen +2T2I Generation Diffusion Language Model Inference
East China Normal University · University of Macau · Zhejiang University +1
Reinforcement Learning for Diffusion LLMs with Entropy-Guided Step Selection and Stepwise Advantages
Mar 13, 2026Vishnu Teja Kunde, Fatemeh Doudi, Mahdi Farahbakhsh +3RL for Diffusion Models Diffusion Language Model Training
Department of Electrical & Computer Engineering, Texas A&M University, College Station, TX 77843
ES-dLLM: Efficient Inference for Diffusion Large Language Models by Early-Skipping
Mar 10, 2026Zijian Zhu, Fei Ren, Zhanhong Tan +1Diffusion Language Model Inference Diffusion Language Models
IIIS, Tsinghua University · Polar Bear Tech.
Latent-DARM: Bridging Discrete Diffusion And Autoregressive Models For Reasoning
Mar 10, 2026Lina Berrayana, Ahmed Heakl, Abdullah Sohail +3Multi-Agent Reasoning Multi-Agent LLMs
EPFL · MBZUAI · ETH Zürich +1
A Comparative analysis of Layer-wise Representational Capacity in AR and Diffusion LLMs
Mar 8, 2026Raghavv Goel, Risheek Garrepalli, Sudhanshu Agrawal +3Diffusion Language Models Representation Geometry in Language Models
Qualcomm AI Research, Qualcomm Technologies Inc.
Diffusion Language Models Are Natively Length-Aware
Mar 6, 2026Vittorio Rossi, Giacomo Cirò, Davide Beltrame +3Adaptive Computation for Diffusion Models Diffusion Language Models
Department of Computing Sciences, Bocconi University, Milan, Italy.
Beyond Scattered Acceptance: Fast and Coherent Inference for DLMs via Longest Stable Prefixes
Mar 5, 2026Pengxiang Li, Joey Tsai, Hongwei Xue +2Diffusion Language Model Inference Diffusion Language Model Decoding
Accio Team, Alibaba Group · Tsinghua University
Free Lunch for Pass@ $k$ ? Low Cost Diverse Sampling for Diffusion Language Models
Mar 5, 2026Sean Lamont, Christian Walder, Paul Montague +2Diversity-Aware Sampling Output Diversity in Language Models
Australian National University · Defence Science and Technology Group · Google DeepMind +1
Progressive Refinement Regulation for Accelerating Diffusion Language Model Decoding
Mar 4, 2026Lipeng Wan, Jianhui Gu, Junjie Ma +4Diffusion Language Model Decoding Diffusion Language Models
Department of Artificial Intelligence, Xi’an Jiaotong University, Xi’an, Shaanxi, China · College of Computing and Data Science, Nanyang Technological University, Singapore · School of Computer Science and Technology, Harbin Institute of Technology, Harbin, China
Beyond Masks: Efficient, Flexible Diffusion Language Models via Deletion-Insertion Processes
Mar 4, 2026Fangyu Ding, Ding Ding, Sijin Chen +8Diffusion Language Models Language Modeling
HKUST · Huawei Foundation Model Dept · CUHK
Efficient Self-Evaluation for Diffusion Language Models via Sequence Regeneration
Mar 3, 2026Linhao Zhong, Linyu Wu, Wen Wang +5Language Model Self-Assessment Confidence Estimation in Language Models
Zhejiang University · National University of Singapore · Zhejiang University of Technology
Characterizing Memorization in Diffusion Language Models: Generalized Extraction and Sampling Effects
Mar 2, 2026Xiaoyu Luo, Wenrui Yu, Qiongxiu Li +1Training Data Extraction Memorization in Language Models
Department of Computer Science, Aalborg University, Copenhagen, Denmark · Department of Electronic Systems, Aalborg University, Copenhagen, Denmark
CRoCoDiL: Continuous and Robust Conditioned Diffusion for Language
Mar 2, 2026Roy Uziel, Omer Belhasin, Itay Levy +4Diffusion Language Models Continuous Diffusion Language Models
NVIDIA
D3LM: A Discrete DNA Diffusion Language Model for Bidirectional DNA Understanding and Generation
Mar 2, 2026Zhao Yang, Hengchang Liu, Chuan Cao +1Molecular Foundation Models Genomic Foundation Models
Gaoling School of Artificial Intelligence, Renmin University of China · Zhongguancun Academy
DUEL: Exact Likelihood for Masked Diffusion via Deterministic Unmasking
Mar 2, 2026Gilad Turok, Chris De Sa, Volodymyr KuleshovMasked Diffusion Models Diffusion Language Models
Department of Computer Science, Cornell University, New York, New York, USA.
MetaState: Persistent Working Memory Enhances Reasoning in Discrete Diffusion Language Models
Mar 2, 2026Kejing Xia, Mingzhe Li, Lixuan Wei +5Diffusion Language Models Persistent Memory for Language Models
Georgia Institute of Technology · University of Massachusetts Amherst · Harvard University
LLaDA-o: An Effective and Length-Adaptive Omni Diffusion Model
Mar 1, 2026Zebin You, Xiaolu Zhang, Jun Zhou +2Diffusion Models Multimodal Diffusion Models
Gaoling School of Artificial Intelligence, Renmin University of China, Beijing, China · Beijing Key Laboratory of Research on Large Models and Intelligent Governance · Engineering Research Center of Next-Generation Intelligent Search and Recommendation, MOE +1
Divide and Conquer: Accelerating Diffusion-Based Large Language Models via Adaptive Parallel Decoding
Feb 27, 2026Xiangzhong Luo, Yilin An, Zhicheng Yu +2Diffusion Language Model Inference Diffusion Language Model Decoding
Southeast University · Nanyang Technological University
Why Diffusion Language Models Struggle with Truly Parallel (Non-Autoregressive) Decoding?
Feb 26, 2026Pengxiang Li, Dilxat Muhtar, Tianlong Chen +2Parallel Decoding for Diffusion Language Models Diffusion Language Model Decoding
The Hong Kong Polytechnic University, Hong Kong, China · ELLIS Institute Tübingen, Tübingen, Germany · Max Planck Institute for Intelligent Systems, Tübingen, Germany +3
Test-Time Scaling with Diffusion Language Models via Reward-Guided Stitching
Feb 26, 2026Roy Miles, Aysim Toker, Andreea-Maria Oncescu +2Structured Reasoning in Language Models Test-Time Scaling
Huawei London Research Center · MVP Lab
Rejection Mixing: Fast Semantic Propagation of Mask Tokens for Efficient DLLM Inference
Feb 26, 2026Yushi Ye, Feng Hong, Huangjie Zheng +4Diffusion Language Model Inference Diffusion Language Model Decoding
Cooperative Medianet Innovation Center, Shanghai Jiao Tong University · School of Artificial Intelligence, Shanghai Jiao Tong University
dLLM: Simple Diffusion Language Modeling
Feb 26, 2026Zhanhui Zhou, Lingjie Chen, Hanghang Tong +1Diffusion Language Model Training Diffusion Language Model Benchmarking
UIUC
The Diffusion Duality, Chapter II: $Ψ$ -Samplers
Feb 24, 2026Justin Deschenaux, Caglar Gulcehre, Subham Sekhar SahooDiffusion Models Diffusion Sampling
EPFL, Lausanne, Switzerland · Microsoft AI · Cornell Tech, NY
Self-Purification Mitigates Backdoors in Multimodal Diffusion Language Models
Feb 24, 2026Guangnian Wan, Qi Li, Gongfan Fang +2Diffusion Language Models Multimodal Diffusion Models
National University of Singapore
Stop-Think-AutoRegress: Language Modeling with Latent Diffusion Planning
Feb 24, 2026Justin Lovelace, Christian Belardi, Sofian Zalouk +3Autoregressive Language Modeling Diffusion Language Models
Department of Computer Science, Cornell University
Adaptation to Intrinsic Dependence in Diffusion Language Models
Feb 23, 2026Yunxiao Zhao, Changxiao CaiDiffusion Language Model Inference Diffusion Language Models
Department of Industrial and Operations Engineering, University of Michigan, Ann Arbor, USA.
Discrete Diffusion Models Exploit Asymmetry to Solve Lookahead Planning Tasks
Feb 23, 2026Itamar Trainin, Shauli Ravfogel, Omri Abend +1Diffusion Language Models Autoregressive Language Modeling
Hebrew University of Jerusalem · New York University
Is Your Diffusion Sampler Actually Correct? A Sampler-Centric Evaluation of Discrete Diffusion Language Models
Feb 23, 2026Luhan Tang, Longxuan Yu, Shaorong Zhang +1Diffusion Language Model Benchmarking Diffusion Sampling
University of California, Riverside

Mar 17, 2026J/K move · Enter open · S save

Re-Mask and Redirect: Exploiting Denoising Irreversibility in Diffusion Language Models

Arth Singh

AIM Intelligence · National Institute of Technology Agartala

Adversarial Attacks on LLMs Adversarial Attacks on Diffusion Models LLM Safety Diffusion Language Models

Open paper AskPDF arXiv

Safety alignment in diffusion language models (dLLMs) relies on a single load-bearing assumption: that committed tokens are permanent. We show that violating this assumption, by re-masking committed refusal tokens and injecting a short affirmative prefix, achieves 74-82% ASR on HarmBench across all three publicly available safety-tuned dLLMs, rising to 92-98% with a generic 8-token compliance prefix. We call this attack TrajHijack; it is the first trajectory-level attack on dLLMs, requires no gradient computation, and generalizes across SFT and preference-optimized (VRPO) models. Three findings emerge. First, the vulnerability is irreducibly two-component: re-masking alone (4.4%) and prefix alone (5.7%) both fail. Second, gradient optimization via a differentiable Gumbel-softmax chain consistently degrades ASR (41.5% vs. 76.1%), because continuous perturbations push token distributions off-manifold. Third, A2D (the strongest published dLLM defense) is more vulnerable to TrajHijack (89.9%) than the undefended model (76.1%): its silent-refusal training removes the contextual resistance that trajectory-level attacks must overcome, an effect we call the Defense Inversion Effect.