Metraipolis › archive › dc_12d36d22d119
HEDGE: Gated Hypergraph Memory for Scalable Multi-Agent Debate — arXiv:2610.38902
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.38902
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T06:16:55.669Z
- SHA-256 of the archived text
4f4f8954fd55520cfba322969935d7490d3beaab413cd6d76b59ba9efa7692b0- Stable URL of this page
https://metraipolis.com/documento/dc_12d36d22d119
How to cite
HEDGE: Gated Hypergraph Memory for Scalable Multi-Agent Debate — arXiv:2610.38902 — arXiv (en), https://arxiv.org/abs/2610.38902 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_12d36d22d119 (Metraipolis archive dc_12d36d22d119, sha256 4f4f8954fd55520cfba322969935d7490d3beaab413cd6d76b59ba9efa7692b0)
BibTeX
@misc{dc12d36d22d119,
title = {HEDGE: Gated Hypergraph Memory for Scalable Multi-Agent Debate — arXiv:2610.38902},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.38902},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_12d36d22d119; sha256 4f4f8954fd55520cfba322969935d7490d3beaab413cd6d76b59ba9efa7692b0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38902. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multi-agent debate systems require agents to retain and revisit arguments across multiple rounds, yet limited context windows and unstructured memories make relevant information increasingly difficult to recover. Existing flat and pairwise-graph memories also struggle to capture higher-order relations, including multi-agent agreement, conflicting evidence, and evolving consensus. We introduce HEDGE, a selective memory framework that represents debate trajectories as semantic hypergraphs. Hyperedges connect related claims across agents, while intra-hyperedge relational graphs preserve fine-grained argumentative structure. A learned memory selection module (MSM), trained through supervised learning and reinforcement learning, routes each memory candidate to STORE, BACKUP, or OMIT, retaining useful information while limiting redundancy. Across seven diverse benchmarks, HEDGE consistently outperforms strong single-agent, multi-agent, and memory baselines, improving accuracy by 5.0 points on average over the strongest multi-agent system with GPT-4o-mini. The gains reach 8.2 points on conflict-heavy tasks, while HEDGE reduces memory usage by 48% and the number of debate rounds by 34%. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2610.38902 — 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 2026-10-31T06:07:23.346Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).