Metraipolis › archive › dc_cd34cc4f75ec

EvoRouter: Cost-Efficient Multi-Agent Routing via Insight-Guided On-Policy Distillation — arXiv:2610.35690

Source: arXiv (en)
Original: https://arxiv.org/abs/2610.35690
License (as archived): CC0-1.0
Archived: 2026-10-27T07:55:29.818Z
SHA-256 of the archived text: 0078de3cfc47c1929ba560feda7f6f2feb977b7457c22a8edda566b8e70a51e4
Stable URL of this page: https://metraipolis.com/documento/dc_cd34cc4f75ec

How to cite

EvoRouter: Cost-Efficient Multi-Agent Routing via Insight-Guided On-Policy Distillation — arXiv:2610.35690 — arXiv (en), https://arxiv.org/abs/2610.35690 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_cd34cc4f75ec (Metraipolis archive dc_cd34cc4f75ec, sha256 0078de3cfc47c1929ba560feda7f6f2feb977b7457c22a8edda566b8e70a51e4)

BibTeX

@misc{dccd34cc4f75ec,
  title = {EvoRouter: Cost-Efficient Multi-Agent Routing via Insight-Guided On-Policy Distillation — arXiv:2610.35690},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.35690},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_cd34cc4f75ec; sha256 0078de3cfc47c1929ba560feda7f6f2feb977b7457c22a8edda566b8e70a51e4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.35690. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Serving web queries at scale with multi-agent systems requires jointly configuring large language models, collaboration modes, agent counts, interaction rounds, and skills. However, the resulting configuration space grows combinatorially, while sparse execution rewards provide limited fine-grained supervision for interdependent planning decisions. We propose EvoRouter, a self-evolving multi-agent routing framework driven by insight-guided on-policy distillation. A dedicated expert periodically synthesizes performance-cost feedback from configuration trajectories across tasks into reusable conditional routing insights about capability requirements, returns from resource scaling, and interactions among configuration choices. A privileged teacher combines these evolving insights with current execution feedback to provide dense token-level supervision for the router's own planning trajectories through on-policy distillation. Cost-aware reinforcement learning jointly optimizes task performance and computational cost. Together, these training signals internalize execution insights into the routing policy. The updated policy generates new trajectories that support subsequent insight extraction, forming an execution-driven self-improvement loop. We evaluate EvoRouter on 19 in-domain datasets spanning five domains and two out-of-domain benchmarks. EvoRouter achieves an average in-domain accuracy of 68.69%, outperforming the strongest routing baseline by 7.23 percentage points while reducing inference compute cost by 50.95%. Our code is open-sourced at this https URL .

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.MA, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI) ; Multiagent Systems (cs.MA).
Pagina del lavoro: https://arxiv.org/abs/2610.35690 — 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 2026-10-27T07:55:15.066Z dalla pagina https://arxiv.org/list/cs.MA/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).