Metraipolis › archive › dc_3d17c7eaf536
When Evidence Shapes Collaboration: Knowledge-Conditioned Topology Generation for Multi-Agent Systems — arXiv:2608.27984
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.27984
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T10:47:23.766Z
- SHA-256 of the archived text
d6e55b35b87704ed6a2635bc921dbc8ef0aa076b7af0f7f8724b2d8c182d7f45- Stable URL of this page
https://metraipolis.com/documento/dc_3d17c7eaf536
How to cite
When Evidence Shapes Collaboration: Knowledge-Conditioned Topology Generation for Multi-Agent Systems — arXiv:2608.27984 — arXiv (en), https://arxiv.org/abs/2608.27984 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3d17c7eaf536 (Metraipolis archive dc_3d17c7eaf536, sha256 d6e55b35b87704ed6a2635bc921dbc8ef0aa076b7af0f7f8724b2d8c182d7f45)
BibTeX
@misc{dc3d17c7eaf536,
title = {When Evidence Shapes Collaboration: Knowledge-Conditioned Topology Generation for Multi-Agent Systems — arXiv:2608.27984},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.27984},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3d17c7eaf536; sha256 d6e55b35b87704ed6a2635bc921dbc8ef0aa076b7af0f7f8724b2d8c182d7f45},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.27984. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-Agent Systems (MAS) have recently moved from static workflows toward dynamically generated collaboration topologies. However, existing topology generation methods rely primarily on the parametric knowledge of large language models, with external search or retrieval used only as a reactive tool rather than an explicit determinant of collaboration structure. This leads to structure-knowledge misalignment, where systems exhibit redundant interactions or insufficient verification in knowledge-intensive tasks. We propose K-GAT (Knowledge-Guided Agent Topology Generator), a neuro-symbolic framework that formulates collaboration topology design as a knowledge-conditioned structure learning problem, integrating external evidence directly into autoregressive graph generation. Extensive experiments on knowledge-intensive benchmarks demonstrate K-GAT's efficiency and effectiveness: notably on the expert-level GPQA dataset, K-GAT outperforms the LLM-Debate baseline by a substantial margin of +15.7% in accuracy, while consuming less than half the computational tokens. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2608.27984 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T10:33:42.381Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).