arXiv cs.CL· Ni Zhang, Haofeng Yuan, Jianan Zhou, Shiji Song, Zhiguang Cao·· 13 小时前AI 评分29
基于 LLM 的外生世界模型用于动态调度
LLM-Based Exogenous World Modeling for Dynamic Scheduling
AI 导读
预训练 LLM 可作为结构化世界模型的外生组件，预测不确定的未来作业到达。研究者提出 Afterstate World-Model Planning（AWMP），通过 afterstate 分解让 LLM 只预测外生到达，并结合信念条件调度策略与短场景 rollout。
正文 · 原文
Abstract:Dynamic scheduling requires dispatching online arrivals while anticipating future jobs, as poor decisions can compound downstream costs under nonlinear tardiness objectives. We show that a pretrained large language model (LLM) can serve as the exogenous component of a structured world model, forecasting uncertain future job arrivals. Specifically, we exploit an afterstate decomposition, which separates known deterministic shop transitions from uncertain exogenous job arrivals, enabling the LLM to forecast only the latter through a compact belief representation. Building on this, we develop Afterstate World-Model Planning (AWMP), which combines an evolved, belief-conditioned dispatching strategy with short scenario-based rollouts. A surprise-triggered refresh mechanism queries the LLM only when newly observed arrivals deviate sufficiently from previous forecast, and online residual self-calibration further conditions the LLM on its recent forecast errors to handle nonstationary arrivals. Across diverse arrival distributions, AWMP reduces weighted tardiness by 6-21% on dynamic job-shop and 22-46% on flexible job-shop benchmarks, relative to the best of eight classical priority-dispatching rules.
| Comments: | 10 pages, 5 figures, 1 table |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.29308 [cs.AI] |
| (or arXiv:2610.29308v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29308 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Haofeng Yuan [view email]
[v1]
Wed, 21 Oct 2026 07:01:25 UTC (303 KB)
来源：arXiv cs.CL · arxiv.org