-
TsinghuaC3I/ZEDA-Qwen3-30B-A3B-Dynamic
Text Generation • 31B • Updated • 36 • 1 -
TsinghuaC3I/ZEDA-GLM-4.7-Flash-Dynamic
Text Generation • 30B • Updated • 24 • 2 -
TsinghuaC3I/ZEDA
Preview • Updated • 150 • 1 -
Post-Trained MoE Can Skip Half Experts via Self-Distillation
Paper • 2605.18643 • Published • 30
AI & ML interests
Large Language Models
Papers
Meta Expert: Cross-Layer Expert Sharing for Storage- and Retrieval-Efficient MoE
A Survey of Reinforcement Learning for Large Reasoning Models
Datasets and Models of UltraMedical
-
UltraMedical: Building Specialized Generalists in Biomedicine
Paper • 2406.03949 • Published -
TsinghuaC3I/UltraMedical
Viewer • Updated • 410k • 6.03k • 32 -
TsinghuaC3I/UltraMedical-Preference
Preview • Updated • 2.25k • 11 -
TsinghuaC3I/Llama-3-8B-UltraMedical
Text Generation • 8B • Updated • 295 • • 23
-
TsinghuaC3I/ZEDA-Qwen3-30B-A3B-Dynamic
Text Generation • 31B • Updated • 36 • 1 -
TsinghuaC3I/ZEDA-GLM-4.7-Flash-Dynamic
Text Generation • 30B • Updated • 24 • 2 -
TsinghuaC3I/ZEDA
Preview • Updated • 150 • 1 -
Post-Trained MoE Can Skip Half Experts via Self-Distillation
Paper • 2605.18643 • Published • 30
Datasets and Models of UltraMedical
-
UltraMedical: Building Specialized Generalists in Biomedicine
Paper • 2406.03949 • Published -
TsinghuaC3I/UltraMedical
Viewer • Updated • 410k • 6.03k • 32 -
TsinghuaC3I/UltraMedical-Preference
Preview • Updated • 2.25k • 11 -
TsinghuaC3I/Llama-3-8B-UltraMedical
Text Generation • 8B • Updated • 295 • • 23