Article
Why Did MiniMax M2 End Up as a Full Attention Model?
MiniMax-AI
• • 80Mufei Li PRO
ml1996
·
AI & ML interests
None yet
Organizations
upvoted an article 7 months ago
upvoted 4
papers 7 months ago
Graph-KV: Breaking Sequence via Injecting Structural Biases into Large Language Models
Paper • 2506.07334 • Published • 1
Measuring Physical-World Privacy Awareness of Large Language Models: An Evaluation Benchmark
Paper • 2510.02356 • Published • 11
RLP: Reinforcement as a Pretraining Objective
Paper • 2510.01265 • Published • 45
Agent Learning via Early Experience
Paper • 2510.08558 • Published • 276