List your company

False Detection Rate Evaluation on HotpotQA

0FDRSENTINELRAG
-0.060.3450.751.155Jun 4, 2026

Evaluation results

MethodLinks
SENTINELRAGLLM Backend=GPT-5-mini2026.060
SENTINELRAGLLM Backend=Qwen-3-8B2026.060
SENTINELRAGLLM Backend=Gemini-3-Flash2026.060
SENTINELRAGLLM Backend=GPT-OSS-20B2026.060
RAG-WMLLM Backend=Qwen-3-8B2026.060
RAG-WMLLM Backend=GPT-5-mini2026.061
RAG-WMLLM Backend=Gemini-3-Flash2026.061
RAG-WMLLM Backend=GPT-OSS-20B2026.061.5