Metraipolis › archive › dc_126de2f42eb3
AlphaBet: Reinforcement Learning for LLM Event Forecasting Agents — arXiv:2610.22684
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.22684
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T08:42:55.726Z
- SHA-256 of the archived text
c2d453d2053c5c0a16710ca1335c59deb1bdf7bfea559d9f208f157d501e2e87- Stable URL of this page
https://metraipolis.com/documento/dc_126de2f42eb3
How to cite
AlphaBet: Reinforcement Learning for LLM Event Forecasting Agents — arXiv:2610.22684 — arXiv (en), https://arxiv.org/abs/2610.22684 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_126de2f42eb3 (Metraipolis archive dc_126de2f42eb3, sha256 c2d453d2053c5c0a16710ca1335c59deb1bdf7bfea559d9f208f157d501e2e87)
BibTeX
@misc{dc126de2f42eb3,
title = {AlphaBet: Reinforcement Learning for LLM Event Forecasting Agents — arXiv:2610.22684},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.22684},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_126de2f42eb3; sha256 c2d453d2053c5c0a16710ca1335c59deb1bdf7bfea559d9f208f157d501e2e87},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.22684. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Forecasting an unresolved event requires finding relevant evidence and deciding how strongly it supports each outcome. Yet outcome-based training of Large Language Model (LLM) forecasters often begins with preassembled evidence at a single prediction date, leaving evidence acquisition outside the learning loop. We introduce AlphaBet, an agentic reinforcement learning (RL) framework that trains a single LLM policy to search, reason, and report probabilities across multiple pre-resolution dates. Each dated forecast is scored by a proper scoring rule against the eventual outcome, allowing one resolved event to supervise decisions under different information states. We decompose forecasting regret into an evidence-acquisition gap and a probability-reporting gap, and show that Group Relative Policy Optimization (GRPO)'s standard-deviation normalization can distort the reporting target under multi-class proper scores. On GLM-4.5-Air, AlphaBet reduces the Brier score from 0.731 to 0.626 on Forecast-Dojo and from 0.612 to 0.528 on FutureX. Controlled ablations attribute these gains to multi-date training and show that they come from improvements in both information acquisition and probability reporting. Our empirical results validate that GRPO's std normalization degrades proper-scoring-rule performance while leaving top-outcome accuracy largely unchanged. Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI). Pagina del lavoro: https://arxiv.org/abs/2610.22684 — 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 2026-10-19T08:39:10.120Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).