List your company

Database Question Answering on RealEHR (test)

55.2AccuracyGATE
-0.4414.00528.4542.895Jun 4, 2026

Evaluation results

MethodLinks
GATEBackbone=GPT-5.4-mini2026.0655.2
ReFoRCEBackbone=GPT-5.4-mini2026.0648.6
ReDelBackbone=GPT-5.4-mini2026.0644.8
ReActBackbone=GPT-5.4-mini2026.0642.9
GATEBackbone=Qwen3-8B2026.065.6
ReFoRCEBackbone=Qwen3-8B2026.063.8
ReDelBackbone=Qwen3-8B2026.062.5
ReActBackbone=Qwen3-8B2026.061.7