Datasets and models in the paper SSL4RL: Revisiting Self-supervised Learning as
Intrinsic Reward for Visual-Language Reasoning
-
PKU-ML/SSL4RL-MMBench-Position-3B
Image-to-Text • 4B • Updated • 6 • 1 -
PKU-ML/SSL4RL-MMBench-Rotation-3B
Image-to-Text • 4B • Updated • 5 -
PKU-ML/SSL4RL-MMBench-Contrastive-3B
Image-to-Text • 4B • Updated • 20 -
PKU-ML/SSL4RL-MMBench-Jigsaw-3B
Image-to-Text • 4B • Updated • 22
