NitroGen：面向通用游戏智能体的开放基础模型
NitroGen: An Open Foundation Model for Generalist Gaming Agents
从公开游戏视频中自动抽取玩家动作，构建互联网规模的视觉-动作对，并用大规模行为克隆训练跨游戏可迁移的视觉-动作基础模型；这套「视频→动作标注→预训练→下游微调」的流程可直接迁移到机器人操作等视觉-动作策略学习场景。
原文依据 an internet-scale video-action dataset constructed by automatically extracting player actions from publicly available gameplay videos
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们提出 NitroGen，一个面向通用游戏智能体的视觉-动作基础模型，基于覆盖 1000 余款游戏、总计 4 万小时的游戏视频训练而成。我们通过三个关键要素扩展具身智能体：1）一个互联网规模的视频-动作数据集，其构建方式是从公开可得的游戏视频中自动抽取玩家动作；2）一个可衡量跨游戏泛化能力的多游戏基准环境；3）一个以大规模行为克隆训练的统一的视觉-动作模型。NitroGen 在多个领域展现出能力，包括 3D 动作游戏中的战斗遭遇、2D 平台游戏中的高精度控制，以及程序化生成世界中的探索。经微调后，预训练能够有效迁移到未见过的游戏，任务成功率相较从零训练的模型最高取得 52% 的相对提升。我们公开数据集、评测套件与模型权重，以推进通用具身智能体的研究。
英文原文摘要 991 字符
We introduce NitroGen, a vision-action foundation model for generalist gaming agents that is trained on 40,000 hours of gameplay videos across more than 1,000 games. We scale embodied agents through three key ingredients: 1) an internet-scale video-action dataset constructed by automatically extracting player actions from publicly available gameplay videos, 2) a multi-game benchmark environment that can measure cross-game generalization, and 3) a unified vision-action model trained with large-scale behavior cloning. NitroGen exhibits competence across diverse domains, including combat encounters in 3D action games, high-precision control in 2D platformers, and exploration in procedurally generated worlds. When fine-tuned, pre-training transfers effectively to unseen games, achieving up to 52% relative improvement in task success rates over models trained from scratch. We release the dataset, evaluation suite, and model weights to advance research on generalist embodied agents.