论文HF Daily Papers · arXiv 2611.1232413 天前
WARC世界模型80次曝光习得新机制,测试时训练提升智能体效率
本文由 AI 根据公开内容整理,可能有误,请以原文为准:How Many Exposures Does a World Model Need? Learning Unseen Game Mechanics at Test Time(HF Daily Papers · arXiv 2611.12324)
核心内容
研究针对具身智能体在陌生环境中快速适应难题,提出WARC世界模型并构建MechEval评测集,涵盖20款游戏中的69种交互机制。该35M参数Vision Transformer在ARC-AGI-3基准上验证,半数新机制仅需约80次曝光即可达到80%以上的状态预测准确率,且在单次游戏全程保持该精度,显著优于无测试时更新的基线方案。
技术上WARC采用全参数测试时训练(Test-Time Training),每次环境状态转移后即时更新权重以适配新规则。在闭环控制中,WARC与固定图探索器组合成WARC-Agent,通过想象规划弥补探索不足,使官方效率得分从0.15跃升至0.46。该方法证明了小参数量模型可通过高频在线学习解决分布外泛化问题,为机器人实时适应提供低算力路径。
关键看点
- 少样本机制学习:WARC模型对半数未见环境机制仅需约80次交互曝光,即可实现80%以上的状态变化预测准确率并全程保持
- 测试时全参更新:35M参数ViT架构支持每次状态转移后全参数微调,使7/7机制达标率远超仅1/7的冻结参数基线
- 闭环效率倍增:WARC-Agent结合固定图探索器,通过测试时训练将ARC-AGI-3公开游戏效率得分从0.15提升至0.46
- 标准化评测构建:推出MechEval数据集标注69种递归机制,按单元格级别量化评估世界模型对新规则的逐次曝光学习效果
工程与产业观察
该工作证实35M量级世界模型通过测试时全参数更新,可在极少交互下掌握新环境动力学,为具身智能体提供低算力在线适应范式。其MechEval评测与WARC架构为机器人Sim-to-Real及非结构化场景部署提供了可复现的少样本学习基准与工程参考。
英文摘要 (Abstract)
World models enable agents to plan and learn from imagined experience, but in an unfamiliar environment they must learn its mechanics from limited interaction. A mechanic is a rule for how an action or an interaction changes part of the environment. Each occurrence of a mechanic that the model observes is one exposure. We ask how many exposures a world model needs to predict the effects of a new environment's mechanics. We address this question on ARC-AGI-3, a benchmark of interactive games defined by their programs. We introduce MechEval, which labels 69 recurring mechanics across 20 games from these programs. It scores each prediction on the cells the mechanic changes at each exposure. As a test bed, we build WARC (World model for ARC-AGI-3), a 35M-parameter vision transformer that is pretrained on other games and updates all of its parameters after each transition. On held-out games, WARC learns to predict four in five mechanics with at least 80% accuracy on the cells they change and keeps that accuracy to the end of the recordings. Half of them reach that accuracy within about 80 exposures, in a single play of the game, and without updates only one in seven does at all. In closed-loop play on the public games, WARC-Agent, which pairs WARC with a fixed graph explorer, triples the explorer's official efficiency score from 0.15 to 0.46, mostly through test-time training. On the hidden games of the Kaggle leaderboard, our entry scores 0.20.
Hugging Face Daily Papers 热度:2 upvotes 作者:Ku Onoda, Haruto Kawano, Shota Takashiro, Kohei Hayashi
外部来源
本页事实对应的公开页面,以原文为准。(点击将离开本站)