研究库 论文知识库

内容库 / 论文

arXiv:2609.33355 · 多模态

Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models

Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models

类型：arxiv
标识：2609.33355
链接：https://arxiv.org/abs/2609.33355
主分类：multimodal
形态：position
TLDR：Masked diffusion language models (MDMs) admit flexible generation orders, making the unmasking strategy an inference decision. Existing methods vary in how they prioritize positions, control parallelism, restrict selection regions, revise predictions, or plan future denoising, yet it remains unclear when these choices should change during generation. We study this question through strategy reversals, where an alternative action becomes preferable to a fixed choice. We organize MDM inference into five axes--score, cardinality, region, commitment, and planning--and define adaptation opportunity
副分类：llm-infra
待LLM分类：否
来源文件：
/inbox/tom/_candidates/2026-10-01-agent-rag-longcontext-candidates.json

相关 · 按语义匹配

笔记批判性精读 · Entropy-Valley "Length-Adaptive Decoding for Masked Diffusion MT" `arXiv:2608.22274` · dLLM 长度自适应解码新锚

笔记 Cool Papers · cs.CL (papers.cool) · RSS 摘要

仓库 modelscope/richdreamer

笔记 LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuning

笔记 Yannic Kilcher (YouTube) · RSS 摘要

仓库 CompVis/stable-diffusion

arXiv

2609.33355

TLDR

Masked diffusion language models (MDMs) admit flexible generation orders, making the unmasking strategy an inference decision. Existing methods vary in how they prioritize positions, control parallelism, restrict selection regions, revise predictions, or plan future denoising, yet it remains unclear when these choices should change during generation. We study this question through strategy reversals, where an alternative action becomes preferable to a fixed choice. We organize MDM inference into five axes--score, cardinality, region, commitment, and planning--and define adaptation opportunity

相关解读

arXiv ↗ 中文全文