Metraipolis › archive › dc_a1efea2c7110
Strategic Competition and Convergence in LLM Leaderboards — arXiv:2611.09780
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.09780
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-09T22:37:30.388Z
- SHA-256 of the archived text
c59a5a41bfabb61b880b3add79a4ac022a58a42fc9c9fe95aa71cb4c56e62823- Stable URL of this page
https://metraipolis.com/documento/dc_a1efea2c7110
How to cite
Strategic Competition and Convergence in LLM Leaderboards — arXiv:2611.09780 — arXiv (en), https://arxiv.org/abs/2611.09780 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a1efea2c7110 (Metraipolis archive dc_a1efea2c7110, sha256 c59a5a41bfabb61b880b3add79a4ac022a58a42fc9c9fe95aa71cb4c56e62823)
BibTeX
@misc{dca1efea2c7110,
title = {Strategic Competition and Convergence in LLM Leaderboards — arXiv:2611.09780},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.09780},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a1efea2c7110; sha256 c59a5a41bfabb61b880b3add79a4ac022a58a42fc9c9fe95aa71cb4c56e62823},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.09780. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Leaderboards are a popular way to compare large language model (LLM) capabilities and can influence how model developers invest resources. We ask when competition on leaderboards leads developers to produce models with similar capability profiles, which we call monoculture, or similar performance levels. We study this by formulating leaderboard competition as a game in which model developers invest resources across capabilities to maximize their win rate. We show that similar production functions can induce similar capability profiles, while comparative advantage can sustain distinct capability profiles. We then study convergence in performance levels and show that diminishing returns can lead to an increasingly crowded competitive frontier. Using data from LM Arena and OpenRouter, we find that developers improve more in categories where they lag competitors, while still retaining category-specific advantages. At the same time, we find that the frontier of models becomes increasingly crowded, alongside increasingly diversified model usage patterns. Our results suggest that increasing competition does not need to produce interchangeable models, but can instead sustain competitive models with distinct strengths. Tema degli esploratori: governance digitale e beni comuni (elenco arXiv cs.CY, nuovi lavori del giorno). Categorie arXiv: Computer Science and Game Theory (cs.GT) ; Computers and Society (cs.CY); Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2611.09780 — 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 2026-11-09T22:32:01.936Z dalla pagina https://arxiv.org/list/cs.CY/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).