Characterizing Language Use in a Collaborative Situated Game
Published in arXiv preprint arXiv:2512.03381, 2025
This work characterizes how language users develop ad-hoc conventions in collaborative situated environments, analyzing a large-scale corpus from a virtual 3D game. The research was conducted at Berkeley Artificial Intelligence Research (BAIR) under the supervision of Prof. Alane Suhr.
Key Contributions:
- Analyzed situated language use in collaborative virtual environments
- Characterized ad-hoc convention formation patterns
- Developed insights into grounded language understanding
Research Area: Natural Language Processing, Grounded Language Understanding, Human-AI Interaction
Recommended citation: N. Tomlin, N. Zhou, E. Fleisig, L. Chen, T. Wright, L. Vinh, L.X. Ma, S. Eisape, T. Du, T. Zhang, A. Koller, A. Suhr. (2025). "Characterizing Language Use in a Collaborative Situated Game." arXiv preprint arXiv:2512.03381.
Download Paper