MM-Eureka models trained via rule-based RL on the new MMK12 dataset achieve competitive or superior multimodal mathematical reasoning performance compared to models like InternVL2.5-78B while trailing slightly behind o1.
• To find the normal vector to the plane A1C1D, we use the cross product of two vectors in the plane: −−→A1D = (1−0,0−0,1−0) = (1 ,0,1), −−→C1D = (0−0,1−0,1−0) = (0 ,1,1)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning
MM-Eureka models trained via rule-based RL on the new MMK12 dataset achieve competitive or superior multimodal mathematical reasoning performance compared to models like InternVL2.5-78B while trailing slightly behind o1.