List your company

Medical Task on MMLU Med

83.8PerformanceNTK-Selector
70.90474.25277.680.948Nov 10, 2025

Evaluation results

MethodLinks
NTK-SelectorModel=QWEN3-8B2025.1183.8
BaseModel=QWEN3-8B2025.1183.4
DSIRModel=QWEN3-8B2025.1182.7
RandomModel=QWEN3-8B2025.1182.6
LESSModel=QWEN3-8B2025.1182.6
TSDSModel=QWEN3-8B2025.1182.3
GradientModel=QWEN3-8B2025.1182.2
Domain-OnlyModel=QWEN3-8B2025.1181.9
EmbeddingModel=QWEN3-8B2025.1181.5
GradientModel=LLAMA3-8B-INSTRUCT2025.1174
NTK-SelectorModel=LLAMA3-8B-INSTRUCT2025.1173.8
LESSModel=LLAMA3-8B-INSTRUCT2025.1173.1
RandomModel=LLAMA3-8B-INSTRUCT2025.1173
EmbeddingModel=LLAMA3-8B-INSTRUCT2025.1172.7
DSIRModel=LLAMA3-8B-INSTRUCT2025.1172.7
BaseModel=LLAMA3-8B-INSTRUCT2025.1172.3
Domain-OnlyModel=LLAMA3-8B-INSTRUCT2025.1171.8
TSDSModel=LLAMA3-8B-INSTRUCT2025.1171.4