Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy
Paper • 2507.01352 • Published • 58
Skywork-Reward-V2
Scaling preference data curation to the extreme
-
-
Skywork/Skywork-Reward-V2-Llama-3.1-8B
Text Classification • 8B • Updated • 25.3k • 41 -
Skywork/Skywork-Reward-V2-Llama-3.1-8B-40M
Text Classification • 8B • Updated • 1.1k • 22 -
Skywork/Skywork-Reward-V2-Llama-3.2-1B
Text Classification • 1B • Updated • 11.3k • 6 -
Skywork/Skywork-Reward-V2-Llama-3.2-3B
Text Classification • 3B • Updated • 324 • 4 -
Skywork/Skywork-Reward-V2-Qwen3-0.6B
Text Classification • 0.6B • Updated • 58k • • 12 -
Skywork/Skywork-Reward-V2-Qwen3-1.7B
Text Classification • 2B • Updated • 1.02k • 7 -
Skywork/Skywork-Reward-V2-Qwen3-4B
Text Classification • 4B • Updated • 1.15k • 10 -
Skywork/Skywork-Reward-V2-Qwen3-8B
Text Classification • 8B • Updated • 4.3k • 22