List your company

Clinical Reasoning on MedR-Bench Treatment

94.59AccuracyGSEM
41.893255.574169.25582.9359Mar 23, 2026

Evaluation results

MethodLinks
GSEMBackbone=DeepSeek-V3.22026.0394.59
A-MemBackbone=DeepSeek-V3.22026.0392.57
ReMeBackbone=DeepSeek-V3.22026.0389.19
FLEXBackbone=DeepSeek-V3.22026.0389.19
GraphRAGBackbone=DeepSeek-V3.22026.0387.16
Naïve RAGBackbone=DeepSeek-V3.22026.0384.46
GSEMBackbone=Qwen3.5-35B2026.0366.89
A-MemBackbone=Qwen3.5-35B2026.0365.54
ReMeBackbone=Qwen3.5-35B2026.0364.86
FLEXBackbone=Qwen3.5-35B2026.0363.51
VanillaBackbone=Qwen3.5-35B2026.0360.14
VanillaBackbone=DeepSeek-V3.22026.0357.43
Mem0Backbone=DeepSeek-V3.22026.0357.43
Mem0gBackbone=DeepSeek-V3.22026.0356.08
Naïve RAGBackbone=Qwen3.5-35B2026.0351.35
GraphRAGBackbone=Qwen3.5-35B2026.0347.3
Mem0Backbone=Qwen3.5-35B2026.0343.92
Mem0gBackbone=Qwen3.5-35B2026.0343.92