Benchmark
SimulatorArena
A dialogue arena benchmark based on LLM-simulated users, designed to compare different agents under identical conditions through interaction with simulated users. The benchmark focuses on cross-agent comparisons to evaluate the quality of natural conversational interactions.
Timeline (last 90 days)
Sep 14
The GAUGE study covered 25 agents from six providers and tested them on two benchmarks: τ²-bench and SimulatorArena.
Unverified50%