MiniMax-01: Scaling Foundation Models with Lightning Attention
Paper
• 2501.08313 • Published
• 300
Attention
-
-
Lizard: An Efficient Linearization Framework for Large Language Models
Paper • 2507.09025 • Published • 19 -
On the Expressiveness of Softmax Attention: A Recurrent Neural Network Perspective
Paper • 2507.23632 • Published • 6 -
Causal Attention with Lookahead Keys
Paper • 2509.07301 • Published • 21 -
Hybrid Architectures for Language Models: Systematic Analysis and Design Insights
Paper • 2510.04800 • Published • 37 -
Less is More: Recursive Reasoning with Tiny Networks
Paper • 2510.04871 • Published • 509 -
Why Low-Precision Transformer Training Fails: An Analysis on Flash Attention
Paper • 2510.04212 • Published • 26 -
Reactive Transformer (RxT) -- Stateful Real-Time Processing for Event-Driven Reactive Language Models
Paper • 2510.03561 • Published • 25 -
Every Attention Matters: An Efficient Hybrid Architecture for Long-Context Reasoning
Paper • 2510.19338 • Published • 115 -
Kimi Linear: An Expressive, Efficient Attention Architecture
Paper • 2510.26692 • Published • 127 -
DoPE: Denoising Rotary Position Embedding
Paper • 2511.09146 • Published • 97 -
SSA: Sparse Sparse Attention by Aligning Full and Sparse Attention Outputs in Feature Space
Paper • 2511.20102 • Published • 28 -
InfiniteVL: Synergizing Linear and Sparse Attention for Highly-Efficient, Unlimited-Input Vision-Language Models
Paper • 2512.08829 • Published • 21 -
MHLA: Restoring Expressivity of Linear Attention via Token-Level Multi-Head
Paper • 2601.07832 • Published • 52 -
Reinforced Attention Learning
Paper • 2602.04884 • Published • 28 -
Prism: Spectral-Aware Block-Sparse Attention
Paper • 2602.08426 • Published • 36 -
SLA2: Sparse-Linear Attention with Learnable Routing and QAT
Paper • 2602.12675 • Published • 53