Post
2654
Skywork-VL Reward🔥A multimodal reward model for both understanding & reasoning tasks, released by Skywork 昆仑万物-天工
Paper: Skywork-VL Reward: An Effective Reward Model for Multimodal Understanding and Reasoning (2505.07263)
Model: Skywork/Skywork-VL-Reward-7B
✨ 7B
✨ Trained on large scale, high-quality preference data
✨ SOTA on VL-RewardBench + boosts reasoning via MPO
Paper: Skywork-VL Reward: An Effective Reward Model for Multimodal Understanding and Reasoning (2505.07263)
Model: Skywork/Skywork-VL-Reward-7B
✨ 7B
✨ Trained on large scale, high-quality preference data
✨ SOTA on VL-RewardBench + boosts reasoning via MPO