arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.05935cs.LGcs.AI
收藏

LLM-FS-Agent:一种用于透明特征选择的基于角色的深思型大语言模型架构

LLM-FS-Agent: A Deliberative Role-based Large Language Model Architecture for Transparent Feature Selection

R&D Department, Audensiel Conseil(Audensiel Conseil 研发部);Ecole Centrale de Lyon(Lyon 工程学院)

Mohamed Bal-Ghaoui, Fayssal Sabri

AI总结:

针对高维数据挑战,提出基于多智能体角色辩论的LLM-FS-Agent架构,实现透明特征选择,在物联网入侵检测数据集上性能优越且训练时间平均减少46%。

摘要:

高维数据在机器学习中仍然是一个普遍存在的挑战,常常损害模型的可解释性和计算效率。虽然大语言模型(LLMs)已显示出通过特征选择进行降维的潜力,但现有的基于LLM的方法往往缺乏结构化的推理和对其决策的透明论证。本文介绍了LLM-FS-Agent,一种新颖的多智能体架构,旨在实现可解释且稳健的特征选择。该系统在多个LLM智能体之间编排了一场深思型的“辩论”,每个智能体被分配一个特定角色,从而能够对特征相关性进行集体评估并生成详细的论证。我们在网络安全领域使用CIC-DIAD 2024物联网入侵检测数据集评估了LLM-FS-Agent,并将其性能与强基线方法(包括LLM-Select和PCA等传统方法)进行了比较。实验结果表明,LLM-FS-Agent consistently在分类性能上达到更优或相当的水平,同时将下游训练时间平均减少了46%(统计显著改进,XGBoost的p值为0.028)。这些发现强调了所提出的深思型架构增强了决策透明度和计算效率,使LLM-FS-Agent成为实际应用中实用且可靠的解决方案。

英文摘要:

High-dimensional data remains a pervasive challenge in machine learning, often undermining model interpretability and computational efficiency. While Large Language Models (LLMs) have shown promise for dimensionality reduction through feature selection, existing LLM-based approaches frequently lack structured reasoning and transparent justification for their decisions. This paper introduces LLM-FS-Agent, a novel multi-agent architecture designed for interpretable and robust feature selection. The system orchestrates a deliberative "debate" among multiple LLM agents, each assigned a specific role, enabling collective evaluation of feature relevance and generation of detailed justifications. We evaluate LLM-FS-Agent in the cybersecurity domain using the CIC-DIAD 2024 IoT intrusion detection dataset and compare its performance against strong baselines, including LLM-Select and traditional methods such as PCA. Experimental results demonstrate that LLM-FS-Agent consistently achieves superior or comparable classification performance while reducing downstream training time by an average of 46% (statistically significant improvement, p = 0.028 for XGBoost). These findings highlight that the proposed deliberative architecture enhances both decision transparency and computational efficiency, establishing LLM-FS-Agent as a practical and reliable solution for real-world applications.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑