arXiv：cs.AI· Tobias Schneider, Danial Fahrner, Soren Antebi, Daniel Uedelhoven, Lennard Bodden, Sandra Halscheidt, Rafet Sifa·· 22 小时前
BehaviorFormer：用可聚类下一 token 预测发现行为模式
BehaviorFormer: Discovering Behavioral Patterns via Clusterable Next-Token Prediction
AI 导读
BehaviorFormer 将下一 token 预测 Transformer 与深度聚类算法 DCSS 结合，学习聚类分离良好的表征。在一场 26 名玩家的受控 Minecraft 会话中，该方法发现 23 种可跨玩家泛化的行为模式，并归为 stationary、roaming、travel、mining、base work、logging、farming 七类。
正文
Abstract:Discovering behavioral patterns in open-ended games is challenging due to the lack of clear objectives or ground-truth annotations. In this work, we extend a next-token prediction Transformer with the deep clustering algorithm DCSS to learn meaningful representations with well-separated clusters. In a controlled Minecraft session with 26 players, the method discovers 23 distinct behavioral patterns that generalize across the studied players, further refining structures present in self-supervised representations into clearly separable clusters. We group the clusters into seven interpretable categories covering stationary, roaming, travel, mining, base work, logging, and farming behavior.
| Comments: | Accepted at IEEE Conference on Games 2026 |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC) |
| Cite as: | arXiv:2610.16322 [cs.LG] |
| (or arXiv:2610.16322v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16322 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Tobias Schneider [view email]
[v1]
Mon, 12 Oct 2026 07:43:23 UTC (384 KB)
来源：arXiv：cs.AI · arxiv.org