---
id: 20261001-T0-12
title: "Grab a Coffee：让离散扩散模型也能用序列级目标引导生成"
title_en: "Grab a Coffee: Future-Aware Guidance for Discrete Diffusion Models"
url: https://ai.daily.yangsir.net/daily/20261001-T0-12
issue_date: 2026-10-01
publish_date: 2026-09-30T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2609.35924
---

# Grab a Coffee：让离散扩散模型也能用序列级目标引导生成

研究者提出Grab a Coffee方法，为离散扩散模型引入未来感知的引导机制。离散扩散模型通过并行解析多个token来生成序列，相比从左到右的生成方式更灵活，但用序列级目标引导这一过程很困难，因为每个token的值取决于尚未确定的后续token。该方法通过编译目标函数，让模型在生成过程中提前感知序列级别的约束，从而在保持并行生成效率的同时实现可控输出。这一思路适用于需要序列级属性控制的离散扩散应用场景，如分子生成和文本生成中的约束满足。

## English Version

**Grab a Coffee: Future-Aware Guidance for Discrete Diffusion Models**

Researchers propose Grab a Coffee, a future-aware guidance method for discrete diffusion models. These models generate sequences by resolving multiple tokens in parallel, offering flexibility over left-to-right generation, but guiding the process with sequence-level objectives is difficult because each token's value depends on future tokens not yet determined. Grab a Coffee compiles the objective so the model can anticipate sequence-level constraints during generation, achieving controllable outputs while maintaining parallel generation efficiency. The approach applies to discrete diffusion tasks requiring sequence-level property control, such as molecule generation and constrained text generation.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2609.35924)

**详情页**：https://ai.daily.yangsir.net/daily/20261001-T0-12

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*