LETGAMESEVAL
Benchmarks
TaskDatasetSOTA resultTrendResults
LETGAMESEVAL Human-based Controls
4.95Helpfulness Score
4
LETGAMESEVAL Human-portrayed SPs
4.74Helpfulness Score
4
LETGAMESEVAL LLM-based Controls
4.93Helpfulness Score
4
LETGAMESEVAL LLM-based SPs
4.98Helpfulness Score
4