研究人员正在开发新方法来改进扩散语言模型（DLM）的解码过程。DLM能够并行生成文本，但目前在质量上落后于自回归模型。几篇论文提出了新颖技术，通过更好地捕捉词元关系和改进扩散解码器与语言模型之间的接口来弥合这一差距。这些进展旨在提高DLM生成的速度和准确性，使其在数学推理和代码生成等复杂任务中更具竞争力。 AI
影响 这些进展可以显著提高并行文本生成的效率和有效性，使扩散模型在复杂的AI应用中更具可行性。
排序理由 多篇学术论文提出改进扩散语言模型解码的新颖方法。
AI 生成摘要 · Google Gemini · 来自 5 个来源。 我们如何撰写摘要 →
arXiv:2606.09159v1 Announce Type: cross Abstract: Diffusion Language Models (DLMs) enable parallel text generation by iteratively denoising a full sequence, offering attractive flexibility compared to auto-regressive (AR) decoding. However, existing methods fail to fully capture …
arXiv:2606.08810v1 Announce Type: cross Abstract: Gaussian-corrupted sentence embeddings have no direct linguistic interpretation, yet continuous diffusion language models can generate fluent text from them. We study this puzzle through Embedded Language Flows (ELF) and identify …
Diffusion Language Models (DLMs) enable parallel text generation by iteratively denoising a full sequence, offering attractive flexibility compared to auto-regressive (AR) decoding. However, existing methods fail to fully capture token relationships, leading to a performance gap …
Block-wise semi-autoregressive decoding is the standard inference paradigm for diffusion large language models (DLMs), but it imposes a strict dependency between blocks: the next block cannot begin until the current block is fully decoded or its denoising budget is exhausted. We …
Diffusion language models (DLMs) offer substantial speed advantages through parallel decoding, but the lack of token dependencies limits generation quality compared to autoregressive (AR) models. Recent progress attempts to bridge the gap via importance sampling, with DLM being t…