Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.07799 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910227169804288 |
|---|---|
| author | Jiang, Eric Hanchen Li, Mengting Wan, Guancheng Yin, Sophia Wu, Yuchen Liang, Xiao Li, Xinfeng Sun, Yizhou Wang, Wei Chang, Kai-Wei Wu, Ying Nian |
| author_facet | Jiang, Eric Hanchen Li, Mengting Wan, Guancheng Yin, Sophia Wu, Yuchen Liang, Xiao Li, Xinfeng Sun, Yizhou Wang, Wei Chang, Kai-Wei Wu, Ying Nian |
| contents | The efficiency of multi-agent systems driven by large language models (LLMs) largely hinges on their communication topology. However, designing an optimal topology is a non-trivial challenge, as it requires balancing competing objectives such as task performance, communication cost, and robustness. Existing frameworks often rely on static or hand-crafted topologies, which inherently fail to adapt to diverse task requirements, leading to either excessive token consumption for simple problems or performance bottlenecks for complex ones. To address this challenge, we introduce a novel generative framework called \textit{Guided Topology Diffusion (GTD)}. Inspired by conditional discrete graph diffusion models, GTD formulates topology synthesis as an iterative construction process. At each step, the generation is steered by a lightweight proxy model that predicts multi-objective rewards (e.g., accuracy, utility, cost), enabling real-time, gradient-free optimization towards task-adaptive topologies. This iterative, guided synthesis process distinguishes GTD from single-step generative frameworks, enabling it to better navigate complex design trade-offs. We validated GTD across multiple benchmarks, and experiments show that this framework can generate highly task-adaptive, sparse, and efficient communication topologies, significantly outperforming existing methods in LLM agent collaboration. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07799 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models Jiang, Eric Hanchen Li, Mengting Wan, Guancheng Yin, Sophia Wu, Yuchen Liang, Xiao Li, Xinfeng Sun, Yizhou Wang, Wei Chang, Kai-Wei Wu, Ying Nian Computation and Language Artificial Intelligence The efficiency of multi-agent systems driven by large language models (LLMs) largely hinges on their communication topology. However, designing an optimal topology is a non-trivial challenge, as it requires balancing competing objectives such as task performance, communication cost, and robustness. Existing frameworks often rely on static or hand-crafted topologies, which inherently fail to adapt to diverse task requirements, leading to either excessive token consumption for simple problems or performance bottlenecks for complex ones. To address this challenge, we introduce a novel generative framework called \textit{Guided Topology Diffusion (GTD)}. Inspired by conditional discrete graph diffusion models, GTD formulates topology synthesis as an iterative construction process. At each step, the generation is steered by a lightweight proxy model that predicts multi-objective rewards (e.g., accuracy, utility, cost), enabling real-time, gradient-free optimization towards task-adaptive topologies. This iterative, guided synthesis process distinguishes GTD from single-step generative frameworks, enabling it to better navigate complex design trade-offs. We validated GTD across multiple benchmarks, and experiments show that this framework can generate highly task-adaptive, sparse, and efficient communication topologies, significantly outperforming existing methods in LLM agent collaboration. |
| title | Dynamic Generation of Multi-LLM Agents Communication Topologies with Graph Diffusion Models |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2510.07799 |