用大模型语言推理自动调优机器学习配置，提升效果与可解释性。
Language-Guided Tuning: Configuration Optimization for Automated ML Research
- 三代理协同：建议、评估、优化，形成自迭代反馈环。
- 在7个数据集上优于传统方法，兼顾性能与可解释性。
- 适合想自动化调参又关心决策逻辑的研究者。
配置优化仍是机器学习中的关键瓶颈，需协调模型架构、训练策略、特征工程和超参数的调优。传统方法独立处理各维度且缺乏可解释性，近期自动化方法则在动态适应性和优化决策的语义理解上表现不足。本文提出语言引导调优（Language-Guided Tuning, LGT），利用多智能体大语言模型通过自然语言推理实现配置自动优化。该框架引入文本反馈信号，补充数值优化，提供对训练动态和配置依赖关系的语义理解。LGT包含三个专用智能体：建议者（Advisor）提出配置调整，评估者（Evaluator）衡量进展，优化者（Optimizer）改进决策过程，构建自我提升的反馈循环。在七个不同数据集上的全面评估表明，LGT显著优于传统优化方法，同时保持高可解释性。
原文摘要 · Abstract (English)
Configuration optimization remains a critical bottleneck in machine learning, requiring coordinated tuning across model architecture, training strategy, feature engineering, and hyperparameters. Traditional approaches treat these dimensions independently and lack interpretability, while recent automated methods struggle with dynamic adaptability and semantic reasoning about optimization decisions. We introduce Language-Guided Tuning (LGT), a framework that employs multi-agent Large Language Models to automatically optimize configurations through natural language reasoning. We apply textual feedback signals that complement numerical optimization by providing semantic understanding of training dynamics and configuration interdependencies. LGT coordinates three specialized agents: an Advisor that proposes configuration changes, an Evaluator that assesses progress, and an Optimizer that refines the decision-making process, creating a self-improving feedback loop. Through comprehensive evaluation on seven diverse datasets, LGT demonstrates substantial improvements over traditional optimization methods while maintaining high interpretability.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。