SAILOR：基于求解器辅助的交互式大语言模型优化恢复
SAILOR: Solver-Assisted Interactive LLM-based Optimization Recovery
- University of Waterloo（滑铁卢大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
SAILOR系统利用求解器辅助交互，通过检测缺失数值并向用户提问，实现优化问题的完整建模，在七个基准上验证了可行性。
AI中文摘要：
优化问题的自然语言描述可能不完整或对求解器所需的数值信息（包括成本、容量、需求、界限和惩罚）含糊不清。语言模型可以将描述转换为代码，但当所需值缺失时，它必须停止或猜测。我们提出了SAILOR，一个概念验证系统，用于检测此类不支持的数值选择，向用户提出有针对性的后续问题，并在返回解决方案之前更新优化模型。问题根据不确定性以及求解器估计的每个缺失值对当前模型的影响强度进行优先级排序。我们使用一个理想化的模拟器（返回真实值）在来自七个掩码基准的1,723个实例上评估了该流水线。各数据集的精确目标值一致性范围从27.0%到87.6%，平均每个实例提出1.4至5.7个问题。这些结果在受控的分支揭示反馈下证明了可行性；它们并未衡量与人类用户的表现或一般结构模型修复。代码可在以下网址获取：this https URL。
英文摘要：
Natural-language descriptions of optimization problems may be incomplete or vague about numerical information that a solver requires, including costs, capacities, demands, bounds, and penalties. A language model can translate the description into code, but when a required value is absent it must either stop or guess. We present SAILOR, a proof-of-concept system that detects such unsupported numerical choices, asks the user targeted follow-up questions, and updates the optimization model before returning a solution. Questions are prioritized using uncertainty and solver-derived estimates of how strongly each missing value affects the current model. We evaluate the pipeline on 1,723 instances from seven masked benchmarks using an idealized simulator that returns ground-truth values. Exact objective-value agreement ranges from 27.0% to 87.6% across datasets, with 1.4--5.7 questions per instance on average. These results establish feasibility under controlled branch-and-reveal feedback; they do not measure performance with human users or general structural model repair. Code is available at: https://github.com/sshaghayeghs/SAILOR.
