Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper
•
2404.08801
•
Published
•
62
LM Architectures
-
-
RecurrentGemma: Moving Past Transformers for Efficient Open Language Models
Paper • 2404.07839 • Published • 40 -
Eagle and Finch: RWKV with Matrix-Valued States and Dynamic Recurrence
Paper • 2404.05892 • Published • 28 -
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Paper • 2312.00752 • Published • 132 -
Multi-Head Mixture-of-Experts
Paper • 2404.15045 • Published • 55 -
Jamba: A Hybrid Transformer-Mamba Language Model
Paper • 2403.19887 • Published • 99 -
KAN: Kolmogorov-Arnold Networks
Paper • 2404.19756 • Published • 102 -
Better & Faster Large Language Models via Multi-token Prediction
Paper • 2404.19737 • Published • 65 -
Contextual Position Encoding: Learning to Count What's Important
Paper • 2405.18719 • Published • 3 -
Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
Paper • 2405.21060 • Published • 60 -
An Image is Worth More Than 16x16 Patches: Exploring Transformers on Individual Pixels
Paper • 2406.09415 • Published • 47 -
Alleviating Distortion in Image Generation via Multi-Resolution Diffusion Models
Paper • 2406.09416 • Published • 28 -
Transformers meet Neural Algorithmic Reasoners
Paper • 2406.09308 • Published • 43 -
Samba: Simple Hybrid State Space Models for Efficient Unlimited Context Language Modeling
Paper • 2406.07522 • Published • 35 -
Explore the Limits of Omni-modal Pretraining at Scale
Paper • 2406.09412 • Published • 10 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 17 -
VideoLLM-online: Online Video Large Language Model for Streaming Video
Paper • 2406.11816 • Published • 20