aiqbench analysis
This systematic mapping study of 377 papers provides the first quantitative baseline for LLM-driven NPCs, revealing that while short-term memory and world-state grounding are common, advanced capabilities like long-term memory and reflection remain rare in current implementations. The analysis highlights significant latency challenges (2.6–15 seconds) and identifies six structural barriers to commercial adoption, offering evidence-based recommendations for developers navigating the transition from scripted to generative game characters.
LLM-based NPCs in Video Games: A Systematic Mapping Study
arXiv:2611.05427v1 Announce Type: cross Abstract: For decades, non-player characters (NPCs) have been driven by rule-based systems whose personalities and interactions are scripted before release. Large language models (LLMs) promise NPCs that converse openly, adapt to new situations, and behave in ways developers did not fully author, prompting a wave of research and commercial experimentation. Yet the field lacks a unified picture of what has been built, how well it works, and at what cost. We address this gap with a systematic mapping study of 377 papers from the academic literature, yielding 42 papers analysed against a main research question and ten sub-questions across four thematic axes. We provide the first quantitative characterisation of component adoption, latency, safety coverage, and reproducibility: short-term memory and world-state grounding are near-universal (60% and 69%), whereas long-term memory (31%), reflection (38%), and retrieval-augmented generation (14%) remain rare, and response times span 2.6 to 15 seconds. Finally, we identify six structural barriers to commercial adoption and offer evidence-based recommendations for developers and researchers.
Abstract:For decades, non-player characters (NPCs) have been driven by rule-based systems whose personalities and interactions are scripted before release. Large language models (LLMs) promise NPCs that converse openly, adapt to new situations, and behave in ways developers did not fully author, prompting a wave of research and commercial experimentation. Yet the field lacks a unified picture of what has been built, how well it works, and at what cost. We address this gap with a systematic mapping study of 377 papers from the academic literature, yielding 42 papers analysed against a main research question and ten sub-questions across four thematic axes. We provide the first quantitative characterisation of component adoption, latency, safety coverage, and reproducibility: short-term memory and world-state grounding are near-universal (60% and 69%), whereas long-term memory (31%), reflection (38%), and retrieval-augmented generation (14%) remain rare, and response times span 2.6 to 15 seconds. Finally, we identify six structural barriers to commercial adoption and offer evidence-based recommendations for developers and researchers.
12 pages. Accepted manuscript. Published in IEEE Transactions on Games (Early Access)
Software Engineering (cs.SE); Artificial Intelligence (cs.AI)
arXiv:2611.05427 [cs.SE]
(or arXiv:2611.05427v1 [cs.SE] for this version)
https://doi.org/10.48550/arXiv.2611.05427 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1109/TG.2026.3733302 Focus to learn more DOI(s) linking to related resources
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.
Full text under the publisher's open licence · 279 words