arXiv：cs.AI· Syamantak Kumar, Jiang Guo, Hassan Hamad, Hideo Kobayashi, Yi Xiang, Yezhou Yang, Yanjun Qi, Daniele Bonadiman, Jiarong Jiang·· 3 小时前
Plan-and-Patch：用扩散语言模型做智能体规划与修复
Plan-and-Patch: Diffusion Language Models for Agentic Planning
AI 导读
Plan-and-Patch 是一个 plan-and-act 框架，让扩散语言模型（dLLM）通过并行去掩码生成类程序的结构化计划，并在保持前后步骤不变的情况下修复指定区域。
正文
Abstract:Planning is increasingly important for long-horizon agents, where successful execution requires coordinating subgoals, tool use, and intermediate outcomes over many steps. Yet assumptions made during planning may be invalidated by the environment, tools may return unexpected results, or actions may fail. Effective agents must therefore not only generate plans, but also revise them. Such revisions often affect only part of a plan, leaving the preceding and subsequent structure intact. Rather than regenerate the entire plan and risk unnecessary changes, repair can regenerate the affected region conditioned on the preserved prefix and suffix. We introduce Plan-and-Patch, a plan-and-act framework in which a diffusion language model (dLLM) generates a structured, program-like plan through parallel unmasking and repairs it by filling in selected regions while keeping the surrounding steps fixed. We compare DreamReasoner-8B and Qwen3-8B as diffusion and autoregressive (AR) planners. On Natural Plan without task-specific training, diffusion (53.7%) achieves nearly twice the plan repair success rate of AR (27.0%). After task-specific training on agentic benchmarks, ALFWorld and TextCraft, the planners achieve similar observed success in plan generation, while diffusion reduces mean plan-generation latency by 39-46% relative to AR. Our results show that Plan-and-Patch provides a framework for faster plan generation and effective plan repair in long-horizon agents.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.10786 [cs.AI] |
| (or arXiv:2610.10786v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.10786 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Syamantak Kumar [view email]
[v1]
Wed, 7 Oct 2026 18:45:45 UTC (50 KB)
来源：arXiv：cs.AI · arxiv.org