List your company

Question Answering on Natural Questions (NQ) (EM Metric)

37.8Exact MatchDGPO
-1.0969.00219.129.198Aug 27, 2025

Evaluation results

MethodLinks
DGPOModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0837.8
Teacher-3BModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B2025.0836.5
SFT → PPOModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0833.8
DistiLLMModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0833.3
SeqKDModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0833.1
KDModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0833.1
TAIDModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0832.5
PPOModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0830.6
GKDModel Family=Qwen 2.5, Teacher Model=Qwen 2.5 3B, Student Model=Qwen 2.5 0.5B2025.0826.6
Student-0.5BModel Family=Qwen 2.5, Student Model=Qwen 2.5 0.5B2025.080.4