[于2026年7月7日提交]

题目：面向专家模型适配的开放式场景推理

查看题为《Open-Ended Scenario Reasoning for Specialist Model Adaptation》的论文PDF，作者 Youcheng Zong 等4人

查看PDF HTML（实验性）

摘要：流程工业积累了经过验证的专家模型，但传感器漂移、原料变化和工况切换导致这些模型在新场景中系统性地退化。收集新的标注数据并重新训练成本高昂，而继续使用原模型会带来持续偏差。现有的适配方法需要利用充足的标注数据修改模型参数，难以在部署系统上快速响应。将大语言模型（LLM）直接用作预测器存在幻觉风险和输出不可控的问题，并且此类预测器无法融合来自现场的非结构化场景知识。针对这些局限性，本文提出了面向专家模型的推理驱动开放适配框架（ROAM），该框架利用LLM的世界知识和推理能力，对冻结的专家模型进行适配以应对未见场景，无需重新训练。ROAM将所有校正约束在一个低维、语义可解释的潜空间中。LLM生成的场景判断与在线观测在一个统一的概率框架下进行融合。风险约束机制在LLM证据不可靠或场景突变时抑制校正，并在证据不足时回退到原始冻结模型。在矿物浓密过程和公开的IndPenSim青霉素发酵数据集上的实验表明，ROAM在主要偏移场景（如隐藏偏移）中将平均绝对误差（MAE）降低20%以上，仅增加839个额外参数，且每次步骤的开销低于0.02毫秒。这些结果表明，LLM推理可以被转化为面向已投用工业模型的一种保守适配信号。

提交历史

来自：Youcheng Zong 先生 查看邮件

[v1] 2026年7月7日周二 11:29:45 UTC (18,164 KB)

[Submitted on 7 Jul 2026]

Title:Open-Ended Scenario Reasoning for Specialist Model Adaptation

View a PDF of the paper titled Open-Ended Scenario Reasoning for Specialist Model Adaptation, by Youcheng Zong and 4 other authors

View PDF HTML (experimental)

Abstract:Process industries have accumulated validated specialist models, yet sensor drift, feedstock variation, and regime switching cause these models to degrade systematically in new scenarios. Collecting new labeled data and retraining is costly, while continuing with the original model incurs persistent bias. Existing adaptation methods require modifying model parameters with sufficient labeled data, making rapid response on deployed systems difficult. Using LLMs as direct predictors risks hallucinations and uncontrollable outputs. Such predictors also cannot incorporate unstructured scenario knowledge from the field. To address these limitations, this article proposes Reasoning-Driven Open Adaptation for Specialist Models (ROAM), a framework that uses LLM world knowledge and reasoning to adapt frozen specialist models to unseen scenarios without retraining. ROAM confines all corrections to a low-dimensional, semantically interpretable latent space. LLM-generated scenario judgments and online observations are fused under a unified probabilistic framework. A risk-constrained mechanism suppresses corrections under unreliable LLM evidence or abrupt scenario shifts and falls back to the original frozen model when evidence is insufficient. Experiments on a mineral thickening process and the public IndPenSim penicillin fermentation dataset show that ROAM reduces MAE by over 20% in major shift settings such as hidden shifts with only 839 additional parameters and under 0.02,ms per-step overhead. These results indicate that LLM reasoning can be turned into a conservative adaptation signal for industrial models already in service.

Submission history

From: Youcheng Zong Mr. [

view email

]

[v1]

Tue, 7 Jul 2026 11:29:45 UTC (18,164 KB)