Benchmarking of LLM Detection: Comparing Two Competing Approaches