arXiv cs.AI· Kuan Zhang, Yukun Chen, Zhihao Yang, Yue Su, Xiangnan Wu, Zirong Chen, Run Luo, Jinkun Hou, Tao Tan, Yinhe Zheng, Yiming Li·· 13 小时前AI 评分48
游戏中的扩展定律：面向具身智能体的多虚拟世界持续预训练研究
Scaling in Games: Continued Pre-Training For Embodied Agents In Diverse Virtual Worlds
AI 导读
一项受控研究用 120 组训练配置（0.8B 至 27B 四种模型规模、50 至 1600 小时六档数据预算、从仅 Minecraft 到多世界混合的五种数据配比）考察持续预训练对具身智能体的影响。
正文 · 原文
Authors:Kuan Zhang, Yukun Chen, Zhihao Yang, Yue Su, Xiangnan Wu, Zirong Chen, Run Luo, Jinkun Hou, Tao Tan, Yinhe Zheng, Yiming Li
Abstract:Scaling data, diversity and model capacity has improved embodied learning in physical environments, where diverse experiences remain grounded in shared physical laws. Whether these gains extend to virtual worlds with different visual styles, dynamics, controls, and objectives remains unclear. We investigate this question through a controlled study of continued pre-training (CPT) across three axes: data volume, diversity, and model size. Our 120 training configurations span four model sizes from 0.8B to 27B, six data budgets from 50 to 1,600 hours, and five mixtures ranging from Minecraft-only data to diverse multi-world corpora. We assess performance and generalization using three-level validation losses and gameplay benchmarks from in-corpus worlds to unseen worlds. Our results show: (1) more data improves in-corpus performance, diversity improves transfer to related worlds with limited gains in distant worlds, and model size improves data efficiency but generalizes worse to distant worlds under single-world training; (2) we observe an early interface phase with broad cross-world benefits and a later world phase with gains narrowed to in-corpus and related worlds, and fit an empirical scaling law relating validation loss to data volume, diversity and model size; and (3) our models are competitive with Minecraft specialists and frontier models in 3D worlds, and perform best in real-time settings. However, the limits of imitation and persistent gaps in distant worlds suggest that building generalist embodied agents requires richer training stages beyond continued pre-training.
| Comments: | Project page: this https URL ; Model: this https URL |
| Subjects: | Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.28698 [cs.AI] |
| (or arXiv:2610.28698v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.28698 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Kuan Zhang [view email]
[v1]
Tue, 20 Oct 2026 11:09:09 UTC (5,280 KB)
来源：arXiv cs.AI · arxiv.org