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 2603.02156cs.NIcs.AI
收藏

6G 推理能有多小?面向 AI 原生网络的微小型语言模型缩放

How Small Can 6G Reason? Scaling Tiny-to-Small Language Models for AI-Native Networks

Department of Computer and Network Engineering, United Arab Emirates University, UAE(计算机与网络工程系,阿联酋大学);Research Institute for Digital Future, Khalifa University, UAE(未来数字研究院,哈利法大学)

Mohamed Amine Ferrag, Abderrahmane Lakas, Merouane Debbah

AI总结:

本文系统研究小型语言模型在 6G 网络语义推理中的缩放行为与部署效率,发现 1.5B-3B 参数的中等模型在确定性稳定性和计算效率间取得最佳平衡。

摘要:

新兴的 6G 愿景,反映在 3GPP、IETF、ETSI、ITU-T 和 O-RAN 联盟的持续标准化工作中,日益将网络描述为 AI 原生系统,其中高层语义推理层运行在标准化的控制和数据平面功能之上。尽管前沿规模的大语言模型(如 Qwen2.5-7B 和 Olmo-3-7B)展示了强大的推理能力,但其计算开销限制了在延迟敏感的边缘原生基础设施中的部署。本文对 AI 原生 6G 系统中用于网络级语义推理的紧凑语言模型的缩放行为和部署效率进行了系统的实证研究。使用 6G-Bench(一个与标准化对齐的基准测试,包含五个能力领域的 30 个决策任务),我们评估了从 135M(SmolLM2-135M)到 7B 参数(Qwen2.5-7B)的模型,包括中等规模架构如 Llama-3.2-1B、Granite-1B 和 Qwen2.5-3B。确定性准确率(pass@1)从 135M 的 0.224 增加到 7B 的 0.707,但缩放增益高度不均匀。在 1 到 1.5B 范围内出现显著的稳定性转变,准确率从 0.373(Llama-3.2-1B)上升到 0.531(Qwen2.5-1.5B),不稳定性差距 Delta_5 从 0.356 缩小到 0.138。超过 3B 参数后,改进减弱(从 3B 到 7B 增加 +0.064)。通过单查询推理性能分析和边缘得分(一种通过延迟和内存占用归一化准确率的指标),我们表明每单位边缘资源的语义可靠性并不随参数数量单调缩放。相反,中等规模模型(约 1.5 到 3B)在确定性稳定性和计算效率之间实现了最有利的平衡,为 AI 原生 6G 架构提供了部署相关的指导。所有脚本和结果公开在 https://this URL。

英文摘要:

Emerging 6G visions, reflected in ongoing standardization efforts within 3GPP, IETF, ETSI, ITU-T, and the O-RAN Alliance, increasingly characterize networks as AI-native systems in which high-level semantic reasoning layers operate above standardized control and data-plane functions. Although frontier-scale large language models (LLMs) such as Qwen2.5-7B and Olmo-3-7B demonstrate strong reasoning capability, their computational footprint limits deployment in latency-sensitive, edge-native infrastructures. This paper presents a systematic empirical study of the scaling behavior and deployment efficiency of compact language models for network-level semantic reasoning in AI-native 6G systems. Using 6G-Bench, a standardization-aligned benchmark comprising 30 decision-making tasks across five capability domains, we evaluate models ranging from 135M (SmolLM2-135M) to 7B parameters (Qwen2.5-7B), including mid-scale architectures such as Llama-3.2-1B, Granite-1B, and Qwen2.5-3B. Deterministic accuracy (pass@1) increases from 0.224 at 135M to 0.707 at 7B, but scaling gains are highly non-uniform. A pronounced stability transition occurs in the 1 to 1.5B range, where accuracy rises from 0.373 (Llama-3.2-1B) to 0.531 (Qwen2.5-1.5B) and the instability gap Delta_5 contracts from 0.356 to 0.138. Beyond 3B parameters, improvements diminish (+0.064 from 3B to 7B). Through single-query inference profiling and an Edge Score metric that normalizes accuracy by latency and memory footprint, we show that semantic reliability per unit edge resource does not scale monotonically with parameter count. Instead, mid-scale models (approximately 1.5 to 3B) achieve the most favorable balance between deterministic stability and computational efficiency, providing deployment-relevant guidance for AI-native 6G architectures. All scripts and results are publicly available at https://github.com/maferrag/6G-Bench

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。

↑