List your company

Health Multimodal Evaluation on HealthBench English (test)

32.7Overall ScorePOW3R dynamic
23.8626.15528.4530.745May 19, 2026

Evaluation results

MethodLinks
POW3R dynamicBase policy=Qwen3-4B, Setting=POW3R dynamic2026.0532.74.75.4
POW3R dynamicBase policy=Qwen3-8B, Setting=POW3R dynamic2026.0531.53.74.6
Category-balancedBase policy=Qwen3-8B, Setting=Category-balanced2026.0530.72.94.8
Category-balancedBase policy=Qwen3-4B, Setting=Category-balanced2026.0530.52.54.6
Static scalarBase policy=Qwen3-8B, Setting=Static scalar2026.0529.92.14.5
Static scalarBase policy=Qwen3-4B, Setting=Static scalar2026.0529.61.63.8
POW3R dynamicBase policy=Gemma3-4B, Setting=POW3R dynamic2026.0528.13.82.7
BaseBase policy=Qwen3-4B, Setting=Base2026.0528—4.2
BinaryBase policy=Qwen3-8B, Setting=Binary2026.0527.90.14.4
BinaryBase policy=Qwen3-4B, Setting=Binary2026.0527.80.24
BaseBase policy=Qwen3-8B, Setting=Base2026.0527.8—4.4
Category-balancedBase policy=Gemma3-4B, Setting=Category-balanced2026.0527.12.82.6
Static scalarBase policy=Gemma3-4B, Setting=Static scalar2026.0525.41.12.8
BaseBase policy=Gemma3-4B, Setting=Base2026.0524.3—2.4
BinaryBase policy=Gemma3-4B, Setting=Binary2026.0524.20.12.2