GRRM: Group Relative Reward Modeling for Machine Translation
Paper
• 2602.14028 • Published
GRRM
Datasets, and model checkpoints of our Group Relative Reward Model (GRRM) framework
-
-
double7/Qwen2.5-7B-GRRM
Text Generation • 8B • Updated • 40 -
double7/Qwen2.5-7B-MT-GRRM-Optimized
Text Generation • 8B • Updated • 36 -
double7/Qwen2.5-7B-MT-GRRM-Optimized-CLA
Text Generation • 8B • Updated • 32 -
double7/Qwen2.5-7B-SQM-GenRM
8B • Updated • 9 -
double7/TowerBlocks-MT-Ranking
Viewer • Updated • 18.8k • 31 -
double7/MT_Ranking_Metric_Test
Viewer • Updated • 8.94k • 12 -
double7/TowerBlocks-MT-CoT-ZhEn
Viewer • Updated • 18.8k • 9