Language-Guided Tuning: Enhancing Numeric Optimization with Textual Feedback

TL;DR

This paper introduces Language-Guided Tuning, a framework that uses large language models and textual feedback to improve configuration optimization in machine learning, enhancing interpretability and adaptability.

Contribution

The paper presents a novel multi-agent LLM-based framework for configuration tuning that incorporates semantic textual feedback, advancing interpretability and dynamic optimization.

Findings

LGT outperforms traditional methods on six datasets.

LGT achieves significant performance improvements.

The framework maintains high interpretability.

Abstract

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 novel framework that employs multi-agent Large Language Models to intelligently optimize configurations through natural language reasoning. We apply textual gradients - qualitative 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 six diverse datasets, LGT demonstrates substantial improvements over traditional optimization methods, achieving performance gains while maintaining high interpretability.