BehaviorFormer: Discovering Behavioral Patterns via Clusterable Next-Token Prediction

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.

Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)

arXiv:2610.16322 [cs.LG]

(or arXiv:2610.16322v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2610.16322 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.