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 2604.03088cs.SEcs.LG
收藏

SkVM:重新审视跨异构LLM的技能语言虚拟机

SkVM: Revisiting Language VM for Skills across Heterogenous LLMs and Harnesses

Shanghai Jiao Tong University(上海交通大学)

Le Chen, Erhu Feng, Yubin Xia, Haibo Chen

AI总结:

本文提出SkVM,通过将技能视为代码并利用LLM作为异构处理器,解决技能在不同平台间执行不一致的问题,提升技能可移植性和效率。

摘要:

LLM代理越来越多地采用技能作为可重用的组合单元。尽管技能在不同代理平台中共享,当前系统将它们视为原始上下文,导致相同技能在不同代理中行为不一致。为解决这一挑战,我们分析了118,000个技能,并借鉴传统编译器设计。我们将技能视为代码,LLM视为异构处理器。为使可移植性成为可行方案,我们将技能的要求分解为一组基本能力,并衡量每个模型-Harness对对支持的程度。基于这些能力配置,我们提出了SkVM,一种用于可移植和高效技能执行的编译和运行时系统。在编译时,SkVM执行基于能力的编译、环境绑定和并发提取。在运行时,SkVM应用JIT代码固化和自适应重新编译以优化性能。我们评估了SkVM在八个不同规模的LLM和三个代理Harness上的表现,涵盖了SkillsBench和代表性技能任务。结果表明,SkVM在不同模型和环境中显著提高了任务完成率,同时减少了高达40%的token消耗。在性能方面,SkVM通过增强并行性实现了高达3.2倍的加速,并通过代码固化实现了19-50倍的延迟降低。

英文摘要:

LLM agents increasingly adopt skills as a reusable unit of composition. While skills are shared across diverse agent platforms, current systems treat them as raw context, causing the same skill to behave inconsistently for different agents. This fragility undermines skill portability and execution efficiency. To address this challenge, we analyze 118,000 skills and draw inspiration from traditional compiler design. We treat skills as code and LLMs as heterogeneous processors. To make portability actionable, we decompose a skill's requirements into a set of primitive capabilities, and measure how well each model-harness pair supports them. Based on these capability profiles, we propose SkVM, a compilation and runtime system designed for portable and efficient skill execution. At compile time, SkVM performs capability-based compilation, environment binding, and concurrency extraction. At runtime, SkVM applies JIT code solidification and adaptive recompilation for performance optimization. We evaluate SkVM across eight LLMs of varying scales and three agent harnesses, covering SkillsBench and representative skill tasks. Results demonstrate that SkVM significantly improves task completion rates across different models and environments while reducing token consumption by up to 40%. In terms of performance, SkVM achieves up to 3.2x speedup with enhanced parallelism, and 19-50x latency reduction through code solidification.

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。

↑