태그: GRPO

2026년 7월 21일

• Aha-Moment
• Chain-of-Thought
• DeepSeek
• Emergent-Behavior
• GRPO
• MoE
• Open-Source
• Pure-RL
• Pure-RL-Training
• R1-Zero
• Reasoning

2026년 7월 21일

• Chain-of-Thought
• DeepSeek
• Distillation
• GRPO
• MoE
• Open-Source
• Pure-RL
• Pure-RL-Training
• R1
• Reasoning
• Self-Verification

2026년 7월 21일

• Chain-of-Thought
• GRPO
• Microsoft
• Phi-4-Reasoning
• Reasoning
• Small-Language-Model
• Synthetic-Data