Salvato in:
Dettagli Bibliografici
Autori principali: Jia, Lianchen, Li, Chaoyang, Houde, Qian, Huang, Tianchi, Liu, Jiangchuan, Sun, Lifeng
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2510.18491
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909861338415104
author Jia, Lianchen
Li, Chaoyang
Houde, Qian
Huang, Tianchi
Liu, Jiangchuan
Sun, Lifeng
author_facet Jia, Lianchen
Li, Chaoyang
Houde, Qian
Huang, Tianchi
Liu, Jiangchuan
Sun, Lifeng
contents Control algorithms in production environments typically require domain experts to tune their parameters and logic for specific scenarios. However, existing research predominantly focuses on algorithmic performance under ideal or default configurations, overlooking the critical aspect of Tuning Potential. To bridge this gap, we introduce Crucible, an agent that employs an LLM-driven, multi-level expert simulation to turn algorithms and defines a formalized metric to quantitatively evaluate their Tuning Potential. We demonstrate Crucible's effectiveness across a wide spectrum of case studies, from classic control tasks to complex computer systems, and validate its findings in a real-world deployment. Our experimental results reveal that Crucible systematically quantifies the tunable space across different algorithms. Furthermore, Crucible provides a new dimension for algorithm analysis and design, which ultimately leads to performance improvements. Our code is available at https://github.com/thu-media/Crucible.
format Preprint
id arxiv_https___arxiv_org_abs_2510_18491
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Crucible: Quantifying the Potential of Control Algorithms through LLM Agents
Jia, Lianchen
Li, Chaoyang
Houde, Qian
Huang, Tianchi
Liu, Jiangchuan
Sun, Lifeng
Artificial Intelligence
Control algorithms in production environments typically require domain experts to tune their parameters and logic for specific scenarios. However, existing research predominantly focuses on algorithmic performance under ideal or default configurations, overlooking the critical aspect of Tuning Potential. To bridge this gap, we introduce Crucible, an agent that employs an LLM-driven, multi-level expert simulation to turn algorithms and defines a formalized metric to quantitatively evaluate their Tuning Potential. We demonstrate Crucible's effectiveness across a wide spectrum of case studies, from classic control tasks to complex computer systems, and validate its findings in a real-world deployment. Our experimental results reveal that Crucible systematically quantifies the tunable space across different algorithms. Furthermore, Crucible provides a new dimension for algorithm analysis and design, which ultimately leads to performance improvements. Our code is available at https://github.com/thu-media/Crucible.
title Crucible: Quantifying the Potential of Control Algorithms through LLM Agents
topic Artificial Intelligence
url https://arxiv.org/abs/2510.18491