TimeROME-DLM enables training-free knowledge editing in masked diffusion language models via temporal causal tracing and low-rank residual edit memory applied at inference time.
hub
Llada-moe: A sparse moe diffusion language model.arXiv preprint arXiv:2509.24389
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
hub tools
citation-role summary
background 2
citation-polarity summary
years
2026 15roles
background 2representative citing papers
Locate-then-edit succeeds at the same early-to-mid MLP locations in masked diffusion models as in autoregressive models, but requires optimization over intermediate partial-mask states to handle multi-token targets.
PSD is a training-free framework that jointly optimizes spatial unmasking and temporal speculative decoding in diffusion LLMs to reach up to 5.5x tokens per forward pass while preserving accuracy comparable to greedy decoding.
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.
DiLaServe improves SLO attainment for diffusion language models by up to 56.6 percentage points and reduces latency by up to 46% with less than 1% accuracy drop via deadline-aware scheduling and dynamic reconfiguration.
WiSP achieves up to 1.95x decode throughput on low-resource MoE serving by dynamically paging reused experts and using MV-WSA to allocate VRAM between experts and KV cache, with the offline policy performing well on both prefill and decode.
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.
dMoE aggregates token expert distributions to block level in dLLMs, cutting unique experts from 69.5 to 14.6, memory by 76-80%, and latency by 1.14-1.66x while retaining 99.11% performance.
Introduces TSPD with a trajectory-feature controller and training-free CE to reduce denoising steps in dLLMs while aiming to preserve quality.
Cola DLM proposes a hierarchical latent diffusion model that learns a text-to-latent mapping, fits a global semantic prior in continuous space with a block-causal DiT, and performs conditional decoding, establishing latent prior modeling as an alternative to token-level autoregressive language model
A recursive sparse MoE framework integrated into diffusion models iteratively refines visual tokens via gated module selection to improve structured reasoning and image generation performance.
Three expert-activation strategies (delayed caching, speculative exploration, limited activation) accelerate MoE diffusion LLMs by up to 2.2x with minimal quality change.
U-turn chains are Markov chains formed by short forward-backward diffusion steps that remain on the learned manifold and, with Metropolis-Hastings, sample from energy-modified targets, exhibiting an ergodicity-breaking transition on fragmented manifolds.
Empirical test shows top-1 argmax concentration has zero precision as collapse warning in DLM LoRA training due to pre-equilibrium saturation while max gradient norm provides usable but family-specific detection on short-horizon runs.
citing papers explorer
-
TimeROME-DLM: Temporal Causal Tracing and Low-Rank Inference-Time Knowledge Editing for Masked Diffusion Language Models
TimeROME-DLM enables training-free knowledge editing in masked diffusion language models via temporal causal tracing and low-rank residual edit memory applied at inference time.
-
Knowledge Editing in Masked Diffusion Language Models
Locate-then-edit succeeds at the same early-to-mid MLP locations in masked diffusion models as in autoregressive models, but requires optimization over intermediate partial-mask states to handle multi-token targets.
-
PSD: Pushing the Pareto Frontier of Diffusion LLMs via Parallel Speculative Decoding
PSD is a training-free framework that jointly optimizes spatial unmasking and temporal speculative decoding in diffusion LLMs to reach up to 5.5x tokens per forward pass while preserving accuracy comparable to greedy decoding.
-
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.
-
DiLaServe: High SLO Attainment Serving for Diffusion Language Models
DiLaServe improves SLO attainment for diffusion language models by up to 56.6 percentage points and reduces latency by up to 46% with less than 1% accuracy drop via deadline-aware scheduling and dynamic reconfiguration.
-
WiSP: A Working-Set View of Mixture-of-Experts Serving on Extremely Low-Resource Hardware
WiSP achieves up to 1.95x decode throughput on low-resource MoE serving by dynamically paging reused experts and using MV-WSA to allocate VRAM between experts and KV cache, with the offline policy performing well on both prefill and decode.
-
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.
-
dMoE: dLLMs with Learnable Block Experts
dMoE aggregates token expert distributions to block level in dLLMs, cutting unique experts from 69.5 to 14.6, memory by 76-80%, and latency by 1.14-1.66x while retaining 99.11% performance.
-
Efficient Diffusion LLMs via Temporal-Spatial Parallel Decoding and Confidence Extrapolation
Introduces TSPD with a trajectory-feature controller and training-free CE to reduce denoising steps in dLLMs while aiming to preserve quality.
-
Continuous Latent Diffusion Language Model
Cola DLM proposes a hierarchical latent diffusion model that learns a text-to-latent mapping, fits a global semantic prior in continuous space with a block-causal DiT, and performs conditional decoding, establishing latent prior modeling as an alternative to token-level autoregressive language model
-
The Thinking Pixel: Recursive Sparse Reasoning in Multimodal Diffusion Latents
A recursive sparse MoE framework integrated into diffusion models iteratively refines visual tokens via gated module selection to improve structured reasoning and image generation performance.
-
TEAM: Temporal-Spatial Consistency Guided Expert Activation for MoE Diffusion Language Model Acceleration
Three expert-activation strategies (delayed caching, speculative exploration, limited activation) accelerate MoE diffusion LLMs by up to 2.2x with minimal quality change.
-
Sampling Data with Chains of Forward-Backward Diffusion Steps
U-turn chains are Markov chains formed by short forward-backward diffusion steps that remain on the learned manifold and, with Metropolis-Hastings, sample from energy-modified targets, exhibiting an ergodicity-breaking transition on fragmented manifolds.
-
When Top-1 Fails: Calibrating LoRA Monitors for Masked Diffusion LMs
Empirical test shows top-1 argmax concentration has zero precision as collapse warning in DLM LoRA training due to pre-equilibrium saturation while max gradient norm provides usable but family-specific detection on short-horizon runs.