List your company

Full Trajectory Prediction on EHRWorld-110K (test)

47.5S@10EHRWorld-14B
33.04436.79740.5544.303Feb 3, 2026

Evaluation results

MethodLinks
EHRWorld-14BSize=14B2026.0247.558.271.626.266.792.590.191.376.5
EHRWorld-8BSize=8B2026.0246.857.771.426.965.893.689.191.376.2
EHRWorld-4BSize=4B2026.024656.670.327.464.993.988.691.275.5
Gemini-3.0-Pro-PreviewSize=-2026.0244.554.368.129.956.548.147.347.757.4
GLM-4.7Size=358B2026.0244.252.566.332.247.638.125.430.548.1
GPT-5.2Size=-2026.0243.55467.434.862.763.54955.361.8
DeepSeek-V3.2Size=671B2026.0241.651.864.93455.450.940.244.955.1
MiniMax-M2.1Size=229B2026.024151.36533.850.247.548.64854.4
Qwen3-235B-A22B-InstructSize=235B2026.0240.951.464.733.946.948.334.340.250.6
Qwen3-30B-A3B-InstructSize=30B2026.0240.350.864.533.841.932.441.536.447.6
GPT-OSS-120BSize=120B2026.0240.250.263.437.248.343.234.638.450
Qwen3-Next-80B-A3B-InstructSize=80B2026.0239.850.16338.153.841.647.144.253.7
Qwen2.5-14B-InstructSize=14B2026.0239.350.765.933.942.335.823.728.545.6
Llama-3.3-70B-InstructSize=70B2026.0238.148.462.536.437.758.143.95050.1
Baichuan-M2-32BSize=32B2026.0237.247.260.141.849.945.638.641.850.6
Qwen3-4B-InstructSize=4B2026.0236.547.160.744.847.34525.832.846.9
MedGemma-27B-ITSize=27B2026.0236.44658.849.643.625.920.522.941.8
Qwen3-8BSize=8B2026.0235.946.36043.746.644.837.340.749.1
MedGemma-4B-ITSize=4B2026.0233.64355.456.143.841.214.221.140.1