LiveCultureBench: a Multi-Agent, Multi-Cultural Benchmark for Large Language Models in Dynamic Social Simulations
This work addresses the need for multi-cultural benchmarks in AI to improve reliability and cultural appropriateness in LLM deployments, though it is incremental in extending existing evaluation frameworks.
The paper tackled the problem of evaluating large language models as autonomous agents in dynamic social simulations, introducing LiveCultureBench to assess task completion and adherence to socio-cultural norms, with results showing metrics for task-norm trade-offs and verifier uncertainty across models and cultural profiles.
Large language models (LLMs) are increasingly deployed as autonomous agents, yet evaluations focus primarily on task success rather than cultural appropriateness or evaluator reliability. We introduce LiveCultureBench, a multi-cultural, dynamic benchmark that embeds LLMs as agents in a simulated town and evaluates them on both task completion and adherence to socio-cultural norms. The simulation models a small city as a location graph with synthetic residents having diverse demographic and cultural profiles. Each episode assigns one resident a daily goal while others provide social context. An LLM-based verifier generates structured judgments on norm violations and task progress, which we aggregate into metrics capturing task-norm trade-offs and verifier uncertainty. Using LiveCultureBench across models and cultural profiles, we study (i) cross-cultural robustness of LLM agents, (ii) how they balance effectiveness against norm sensitivity, and (iii) when LLM-as-a-judge evaluation is reliable for automated benchmarking versus when human oversight is needed.