Metraipolis › archive › dc_70eb832790af
Beyond Solo and Consistency: Vindicating Multi-Agent Debate via Conditional Progressive Pruning — arXiv:2609.33974
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.33974
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T21:53:15.508Z
- SHA-256 of the archived text
3e7889988f6e6feb0862e5375bd1b90ef5066c5c9e2daeaf734365f77fe1532b- Stable URL of this page
https://metraipolis.com/documento/dc_70eb832790af
How to cite
Beyond Solo and Consistency: Vindicating Multi-Agent Debate via Conditional Progressive Pruning — arXiv:2609.33974 — arXiv (en), https://arxiv.org/abs/2609.33974 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_70eb832790af (Metraipolis archive dc_70eb832790af, sha256 3e7889988f6e6feb0862e5375bd1b90ef5066c5c9e2daeaf734365f77fe1532b)
BibTeX
@misc{dc70eb832790af,
title = {Beyond Solo and Consistency: Vindicating Multi-Agent Debate via Conditional Progressive Pruning — arXiv:2609.33974},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.33974},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_70eb832790af; sha256 3e7889988f6e6feb0862e5375bd1b90ef5066c5c9e2daeaf734365f77fe1532b},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33974. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large Language Model (LLM) based Multi-Agent Debate (MAD) is one of the most effective test time scaling techniques. Through multi-round communication, agents complement each other in knowledge and reasoning and solve tasks that no single member can solve. However, existing MAD frameworks fail to beat strong Single Agent and Consistency-based baselines under the same strict cost limit, which shakes the foundation of the MAD field. We propose Conditional Progressive Pruning (CPP), a lightweight pruning framework that fully exploits multi-round MAD. CPP outperforms all existing MAD frameworks on multiple dominated benchmarks. It is also the first to fully outperform consistency methods. Our code, detailed agent interaction records will be released soon. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.33974 — 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 2026-10-07T21:27:08.574Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).