HiVA: Self-organized Hierarchical Variable Agent via Goal-driven Semantic-Topological Evolution
This addresses the problem of autonomous agents needing adaptable and efficient task execution for AI researchers and developers, representing a novel method for a known bottleneck rather than an incremental advance.
The paper tackles the trade-off between reusable fixed workflows and flexible reactive loops in autonomous agents by introducing HiVA, a framework that models agentic workflows as self-organized graphs using the Semantic-Topological Evolution algorithm, resulting in 5-10% improvements in task accuracy and enhanced resource efficiency across various benchmarks.
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.