Self-Evolving Multi-Agent Systems via Textual Backpropagation
MCML Authors
Volker Tresp
Prof. Dr.
Core PI
Hinrich Schütze
Prof. Dr.
Core PI
Abstract
Volker Tresp
Prof. Dr.
Core PI
Hinrich Schütze
Prof. Dr.
Core PI
Abstract
Leveraging multiple Large Language Models (LLMs) has proven effective for addressing complex, high-dimensional tasks, but current approaches often rely on static, manually engineered multi-agent configurations. To overcome these constraints, we present the Agentic Neural Network (ANN), a framework that conceptualizes multi-agent collaboration as a layered neural network architecture. In this design, each agent operates as a node, and each layer forms a cooperative team focused on a specific subtask. The proposed framework follows a two-phase optimization strategy: (1) Forward Phase - Drawing inspiration from neural network forward passes, tasks are dynamically decomposed into subtasks, and cooperative agent teams with suitable aggregation methods are constructed layer by layer. (2) Backward Phase - Mirroring backpropagation, we refine both global and local collaboration through iterative feedback, allowing agents to self-evolve their roles, prompts, and coordination. This neuro-symbolic approach enables our framework to create new or specialized agent teams post-training, delivering notable gains in accuracy and adaptability. Across seven benchmark datasets, ANN surpasses leading multi-agent baselines under the same configurations, showing consistent performance improvements.
inproceedings MLZ+26
Findings @ACL 2026
Findings at the 64th Annual Meeting of the Association for Computational Linguistics. San Diego, CA, USA, Jul 02-07, 2026.
Authors
X. Ma • Y. Ma • C. Lin • S. Yan • J. Bi • Z. Cao • Y. Tian • V. Tresp • H. SchützeLinks
DOIResearch Areas
BibTeXKey: MLZ+26