arXiv Machine Learning
Fork-dLLM: Avoiding the Flexibility Trap in Diffusion Language Models
Read the original on arXiv Machine Learning →The Flow has not summarised this story yet — read it at arXiv Machine Learning.
Related stories
arXiv Machine Learning
Aug 12
A Tale of Two Temperatures: Simple, Efficient, and Diverse Sampling from Diffusion Language Models
arXiv:2604. 09921v2 Announce Type: replace Abstract: Much work has been done on designing fast and accurate sampling for diffusion language models (dLLMs).
arXiv AI
Jun 3
$R^2$-dLLM: Accelerating Diffusion Large Language Models via Spatio-Temporal Redundancy Reduction
arXiv:2604. 18995v2 Announce Type: replace-cross Abstract: Diffusion Large Language Models (dLLMs) have emerged as a promising alternative to autoregressive generation by enabling parallel token prediction.
arXiv AI
Jun 2
BlockBatch: Multi-Scale Consensus Decoding for Efficient Diffusion Language Model Inference
arXiv:2605. 29233v2 Announce Type: replace-cross Abstract: Diffusion language models (dLLMs) generate text by iteratively denoising multiple token positions in parallel, offering an attractive alternative to strictly autoregressive decoding.
arXiv Machine Learning
Sep 25
Enabling Approximate Joint Sampling in Diffusion LMs
The paper introduces a lightweight single‑layer sampler that allows masked diffusion language models to approximate joint sampling of multiple tokens in a single full‑model forward pass. By training the sampler to mimic exact joint sampling from a frozen diffusion model, the authors enable parallel unmasking of tokens while maintaining a close match to the true joint distribution. Experiments on Dream‑7B and Llada‑7B models show that unmasking four tokens per denoising step yields a MAUVE score of 0.87, a substantial improvement over the marginal baseline of 0.31.
arXiv Machine Learning
Jun 3
Learning Unmasking Policies for Diffusion Language Models
arXiv:2512. 09106v4 Announce Type: replace Abstract: Diffusion (Large) Language Models (dLLMs) now match the downstream performance of their autoregressive counterparts on many tasks, while holding the promise of being more efficient during inference.
arXiv AI
Sep 16
Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling
arXiv:2609.16450v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer a promising parallel decoding paradigm as an alternative to autoregressive generation through iterative...