D^2SD uses two diffusion drafters in a prefix tree structure with confidence scores to select and recover alternative draft sequences, achieving higher acceptance rates in speculative decoding.
hub
LLaDA2.1 : Speeding up text diffusion via token editing
20 Pith papers cite this work. Polarity classification is still indexing.
20
Pith papers citing it
hub tools
citation-role summary
background 4
citation-polarity summary
years
2026 20roles
background 4polarities
background 4representative citing papers
Infilling extraction on diffusion language models extracts up to three times more verbatim sequences than prefix methods and achieves higher recall on redacted emails than autoregressive models.
Introduces Block-R1 benchmark, Block-R1-41K dataset, and a conflict score to handle domain-specific optimal block sizes in RL post-training of diffusion LLMs.
TAD improves the accuracy-parallelism trade-off in diffusion LLMs via temporal-aware self-distillation that applies hard labels to soon-to-be-decoded tokens and soft supervision to future tokens.
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.
MemDLM embeds a simulated denoising trajectory into DLM training via bi-level optimization, creating a parametric memory that improves convergence and long-context performance even when the memory is dropped at test time.
Joint AR–diffusion training yields one tri-mode LM that switches AR, diffusion, and self-speculation, beating open AR/diffusion models on accuracy and tokens-per-forward.
Introduces RSI metric and RSI-S filtering method for adaptive token selection in RLVR, reporting 2-3 point gains over GRPO on AIME/AMC benchmarks.
Masked discrete diffusion with token editing and grouped cross-entropy reaches strong text-to-image generation scores in an 8B decoder-only model, reporting GenEval 0.90, DPG 86.9, HPSv3 10.76.
MBD-LMs raise average tokens per forward pass from 3.47 to 6.19 (and to 9.34 with DMax) via multi-block teacher forcing and optimized parallel decoding while holding or slightly improving accuracy on math and code tasks.
VoidPadding decouples padding from termination in MDLMs via a new [VOID] token, delivering +17.84 average benchmark points and 55.7% fewer decoding steps on Dream-7B-Instruct.
Self-generated T2T training on LLaDA2.1-mini improves benchmark accuracy and lowers edit intensity by supervising recovery from model-generated corruptions instead of random ones.
MIR improves validation loss in repeated-data pretraining and SoftQ fits data-constrained scaling experiments better than additive laws, equating MIR gains to roughly 1.3 times more unique data.
GDSD reduces RL for dLLMs to likelihood-free self-distillation via a normalization-free logit-matching objective, outperforming ELBO methods with more stable training on LLaDA-8B and Dream-7B.
D²-Monitor routes between lightweight and heavy safety probes using the count of hesitation steps in diffusion LLM denoising trajectories, achieving SOTA trade-off on three datasets with under 0.85M parameters.
TrajDLM applies block diffusion language models to discrete road-segment sequences with topology constraints to generate realistic trajectories up to 2.8 times faster than prior methods while supporting zero-shot transfer.
ReflectDrive-2 combines masked discrete diffusion with RL-aligned self-editing to generate and refine driving trajectories, reaching 91.0 PDMS on NAVSIM camera-only and 94.8 in best-of-6.
OALMs exhibit order-dependent likelihoods up to 0.49 nats/token and a uniform confidence spread maximizes recoverability, motivating Var(log q_t) as a decoding diagnostic.
Re-evaluation finds post-hoc remasking (WINO) yields little-to-no gain over confidence unmasking in standard dLLM settings and can worsen diversity collapse under stochastic decoding.
citing papers explorer
-
D^2SD: Accelerating Speculative Decoding with Dual Diffusion Draft Models
D^2SD uses two diffusion drafters in a prefix tree structure with confidence scores to select and recover alternative draft sequences, achieving higher acceptance rates in speculative decoding.
-
Extracting Training Data from Diffusion Language Models via Infilling
Infilling extraction on diffusion language models extracts up to three times more verbatim sequences than prefix methods and achieves higher recall on redacted emails than autoregressive models.
-
Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models
Introduces Block-R1 benchmark, Block-R1-41K dataset, and a conflict score to handle domain-specific optimal block sizes in RL post-training of diffusion LLMs.
-
TAD: Temporal-Aware Trajectory Self-Distillation for Fast and Accurate Diffusion LLM
TAD improves the accuracy-parallelism trade-off in diffusion LLMs via temporal-aware self-distillation that applies hard labels to soon-to-be-decoded tokens and soft supervision to future tokens.
-
DMax: Aggressive Parallel Decoding for dLLMs
DMax uses On-Policy Uniform Training and Soft Parallel Decoding to enable aggressive parallelism in dLLMs, raising TPF on GSM8K from 2.04 to 5.47 and on MBPP from 2.71 to 5.86 while preserving accuracy.
-
MemDLM: Memory-Enhanced DLM Training
MemDLM embeds a simulated denoising trajectory into DLM training via bi-level optimization, creating a parametric memory that improves convergence and long-context performance even when the memory is dropped at test time.
-
Nemotron-Labs-Diffusion: A Tri-Mode Language Model Unifying Autoregressive, Diffusion, and Self-Speculation Decoding
Joint AR–diffusion training yields one tri-mode LM that switches AR, diffusion, and self-speculation, beating open AR/diffusion models on accuracy and tokens-per-forward.
-
Which Tokens Matter? Adaptive Token Selection for RLVR with the Relative Surprisal Index
Introduces RSI metric and RSI-S filtering method for adaptive token selection in RLVR, reporting 2-3 point gains over GRPO on AIME/AMC benchmarks.
-
Nemotron-Labs-Diffusion-Image: Advancing Masked Discrete Diffusion for High-Resolution Image Synthesis
Masked discrete diffusion with token editing and grouped cross-entropy reaches strong text-to-image generation scores in an 8B decoder-only model, reporting GenEval 0.90, DPG 86.9, HPSv3 10.76.
-
Multi-Block Diffusion Language Models
MBD-LMs raise average tokens per forward pass from 3.47 to 6.19 (and to 9.34 with DMax) via multi-block teacher forcing and optimized parallel decoding while holding or slightly improving accuracy on math and code tasks.
-
VoidPadding: Let [VOID] Handle Padding in Masked Diffusion Language Models so that [EOS] Can Focus on Semantic Termination
VoidPadding decouples padding from termination in MDLMs via a new [VOID] token, delivering +17.84 average benchmark points and 55.7% fewer decoding steps on Dream-7B-Instruct.
-
Self-Generated Error Training for Token Editing in Diffusion Language Models
Self-generated T2T training on LLaDA2.1-mini improves benchmark accuracy and lowers edit intensity by supervising recovery from model-generated corruptions instead of random ones.
-
Data-Constrained Language Model Pretraining: Improved Regularization and Scaling Laws
MIR improves validation loss in repeated-data pretraining and SoftQ fits data-constrained scaling experiments better than additive laws, equating MIR gains to roughly 1.3 times more unique data.
-
GDSD: Reinforcement Learning as Guided Denoiser Self-Distillation for Diffusion Language Models
GDSD reduces RL for dLLMs to likelihood-free self-distillation via a normalization-free logit-matching objective, outperforming ELBO methods with more stable training on LLaDA-8B and Dream-7B.
-
$D^2$-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation-Aware Routing
D²-Monitor routes between lightweight and heavy safety probes using the count of hesitation steps in diffusion LLM denoising trajectories, achieving SOTA trade-off on three datasets with under 0.85M parameters.
-
TrajDLM: Topology-Aware Block Diffusion Language Model for Trajectory Generation
TrajDLM applies block diffusion language models to discrete road-segment sequences with topology constraints to generate realistic trajectories up to 2.8 times faster than prior methods while supporting zero-shot transfer.
-
ReflectDrive-2: Reinforcement-Learning-Aligned Self-Editing for Discrete Diffusion Driving
ReflectDrive-2 combines masked discrete diffusion with RL-aligned self-editing to generate and refine driving trajectories, reaching 91.0 PDMS on NAVSIM camera-only and 94.8 in best-of-6.
-
Decoding in Order-Agnostic Language Models: Chain-Rule Deviation and Uniform Spreading
OALMs exhibit order-dependent likelihoods up to 0.49 nats/token and a uniform confidence spread maximizes recoverability, motivating Var(log q_t) as a decoding diagnostic.
-
Re-evaluating Confidence Remasking in Masked Diffusion Language Models
Re-evaluation finds post-hoc remasking (WINO) yields little-to-no gain over confidence unmasking in standard dLLM settings and can worsen diversity collapse under stochastic decoding.
- Remask, Don't Replace: Token-to-Mask Refinement in Diffusion Language Models