GRPO-MA: Multi-Answer Generation in GRPO for Stable and Efficient Chain-of-Thought Training
Paper
•
2509.24494
•
Published
•
10
•
2
Guanghui Ren
sundrops
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
2 days ago
Real2Edit2Real: Generating Robotic Demonstrations via a 3D Control Interface
upvoted
a
paper
about 2 months ago
FARMER: Flow AutoRegressive Transformer over Pixels
authored
a paper
3 months ago
GRPO-MA: Multi-Answer Generation in GRPO for Stable and Efficient
Chain-of-Thought Training
Organizations
None yet
commented
a
paper
3 months ago
commented
a
paper
5 months ago