Rohan Paul (@rohanpaul_ai) · 2026/10/11 21:02:00
清华 AAArena 基准揭示：AI 智能体通过复盘比赛录像重写代码可登顶游戏榜，但复杂规则下易陷入停滞
清华大学发布 AAArena 基准测试，评估 AI 编码智能体在长期运行的游戏代理竞赛中的启发式学习能力。研究发现，基于详细比赛录像的反馈机制显著优于仅依赖胜负结果，使 Pacman 机器人排名从第 11 位跃升至第 1 位；然而，增加比赛预算并不能帮助卡在瓶颈期的智能体突破复杂规则游戏的限制。
报道全文原始报道全文
Rohan Paul (@rohanpaul_ai) · 2026-10-11 13
New Tsinghua paper finds that AI agents improving game bots from match replays can top human leaderboards, but mostly stall on games with complex rules.
Getting AI to learn a winning game strategy from a limited number of matches is still hard, especially against changing rivals.
They built AAArena from 12 games in Tsinghua's yearly bot-building contest, with 1,920 archived human programs as rivals. A coding agent, with its model weights unchanged, reads the rules, picks opponents, studies replays, and rewrites its bot within a match budget.
Detailed replays beat win/loss-only feedback in all 3 games tested. With replays, a Pacman bot reached rank 1, versus rank 11 without them.
Tripling the match budget did not push any of 4 stuck bots to rank 1.
– arxiv. org/abs/2610.12341
Title: "Can AI Agents Learn Their Way to the Top? Evaluating Heuristic Learning in a Long-Running Game Agent Competition"
