1
FIFA World Cup 2026 as a Contamination-Free Benchmark for LLM Forecasting Agents: Four Models, a Bookmaker, and 104 Matches
用2026年世界杯104场比赛检验四大AI模型的预测能力，前沿研究+实际体育数据评估LLM自主预测代理表现。
arXiv:2607.17765v1 Announce Type: cross Abstract: We introduce WC2026-Agents, a benchmark and dataset for evaluating large language models (LLMs) as a…