Metraipolis › archive › dc_c865f5eb7cdc

JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making — arXiv:2610.19589

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.19589
License (as archived)
CC0-1.0
Archived
2026-10-15T20:01:50.996Z
SHA-256 of the archived text
a0ba9ef64baf55af0a27ffa5fd2a3f4232d5273c14736c60f32f09f27e9778f1
Stable URL of this page
https://metraipolis.com/documento/dc_c865f5eb7cdc

How to cite

JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making — arXiv:2610.19589 — arXiv (en), https://arxiv.org/abs/2610.19589 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c865f5eb7cdc (Metraipolis archive dc_c865f5eb7cdc, sha256 a0ba9ef64baf55af0a27ffa5fd2a3f4232d5273c14736c60f32f09f27e9778f1)
BibTeX
@misc{dcc865f5eb7cdc,
  title = {JockeyBench: Evaluating LLM Agents on Horse-Racing Decision-Making — arXiv:2610.19589},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.19589},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c865f5eb7cdc; sha256 a0ba9ef64baf55af0a27ffa5fd2a3f4232d5273c14736c60f32f09f27e9778f1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.19589. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Repeated event forecasting requires extracting predictive signals from noisy evidence and evaluating their value across successive decisions. We introduce JockeyBench, a benchmark and evaluation protocol that studies this problem through historical Japanese horse racing. The protocol combines time-gated evidence, priced tools, paired frozen-program controls, and two workflows: executing an LLM-written predictor or acquiring evidence for a fresh per-race LLM forecast. Evaluation separates forecast quality, benchmark-specific net utility, and retrospective betting returns and risk under fixed decision rules. Across 28 model--scaffold configurations on 324 races, paired tests identify seven revision gains and two declines after multiple-testing correction. Yet every configuration has negative mean profit under capped-Kelly settlement. In a four-model comparison, the reusable-program workflow has higher mean betting profit and lower research fees, although both workflows lose on average. Further diagnostics reveal that mean probability bias reverses on selected bets and that profitable individual runs can depend on concentrated returns. These findings motivate evaluating predictive competence, adaptation, resource use, and decision value separately when assessing agents for repeated event forecasting.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.19589 — 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 2026-10-15T19:35:48.251Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).