HiVA：通过目标驱动的语义-拓扑演化实现自组织分层可变智能体
HiVA: Self-organized Hierarchical Variable Agent via Goal-driven Semantic-Topological Evolution
AI总结：
针对现有自主智能体范式在固定工作流灵活性不足与反应式结构难迁移间的权衡问题，提出HiVA框架与STEV算法，通过自组织图演化实现未知环境优化，在多基准上准确率提升5-10%且资源效率更高。
摘要：
自主智能体在推动通用人工智能发展中发挥着关键作用，可通过大语言模型（LLM）实现问题分解与工具编排。然而现有范式面临关键权衡：一方面，可复用的固定工作流在环境变化时需要人工重新配置；另一方面，灵活的反应式循环无法将推理过程提炼为可迁移的结构。我们提出分层可变智能体（Hierarchical Variable Agent，HiVA）这一新颖框架，将智能体工作流建模为自组织图，并采用语义-拓扑演化（Semantic-Topological Evolution，STEV）算法，该算法以文本梯度作为离散域反向传播的代理，优化混合语义-拓扑空间。其迭代过程包括融合多臂老虎机的前向路由、基于环境反馈的诊断梯度生成，以及协同更新以共同演化个体语义与拓扑，实现未知环境下的集体优化。在对话、编程、长上下文问答、数学及智能体基准上的实验表明，相比现有基线，该方法任务准确率提升5-10%，资源效率也有所提高，证实了HiVA在自主任务执行中的有效性。
英文摘要：
Autonomous agents play a crucial role in advancing Artificial General Intelligence, enabling problem decomposition and tool orchestration through Large Language Models (LLMs). However, existing paradigms face a critical trade-off. On one hand, reusable fixed workflows require manual reconfiguration upon environmental changes; on the other hand, flexible reactive loops fail to distill reasoning progress into transferable structures. We introduce Hierarchical Variable Agent (HiVA), a novel framework modeling agentic workflows as self-organized graphs with the Semantic-Topological Evolution (STEV) algorithm, which optimizes hybrid semantic-topological spaces using textual gradients as discrete-domain surrogates for backpropagation. The iterative process comprises Multi-Armed Bandit-infused forward routing, diagnostic gradient generation from environmental feedback, and coordinated updates that co-evolve individual semantics and topology for collective optimization in unknown environments. Experiments on dialogue, coding, Long-context Q&A, mathematical, and agentic benchmarks demonstrate improvements of 5-10% in task accuracy and enhanced resource efficiency over existing baselines, establishing HiVA's effectiveness in autonomous task execution.
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。
