-
Mixture-of-Depths: Dynamically allocating compute in transformer-based language models
Paper • 2404.02258 • Published • 103 -
Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs
Paper • 2403.20041 • Published • 34 -
ViTAR: Vision Transformer with Any Resolution
Paper • 2403.18361 • Published • 51
Collections
Discover the best community collections!
Collections including paper arxiv:2403.20041
-
Simple and Scalable Strategies to Continually Pre-train Large Language Models
Paper • 2403.08763 • Published • 48 -
Jamba: A Hybrid Transformer-Mamba Language Model
Paper • 2403.19887 • Published • 102 -
Transformer-Lite: High-efficiency Deployment of Large Language Models on Mobile Phone GPUs
Paper • 2403.20041 • Published • 34 -
Advancing LLM Reasoning Generalists with Preference Trees
Paper • 2404.02078 • Published • 43
-
AtP*: An efficient and scalable method for localizing LLM behaviour to components
Paper • 2403.00745 • Published • 11 -
The Era of 1-bit LLMs: All Large Language Models are in 1.58 Bits
Paper • 2402.17764 • Published • 589 -
MobiLlama: Towards Accurate and Lightweight Fully Transparent GPT
Paper • 2402.16840 • Published • 23 -
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
Paper • 2402.13753 • Published • 110
-
Beyond Language Models: Byte Models are Digital World Simulators
Paper • 2402.19155 • Published • 49 -
Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models
Paper • 2402.19427 • Published • 52 -
VisionLLaMA: A Unified LLaMA Interface for Vision Tasks
Paper • 2403.00522 • Published • 44 -
Resonance RoPE: Improving Context Length Generalization of Large Language Models
Paper • 2403.00071 • Published • 22
-
Griffin: Mixing Gated Linear Recurrences with Local Attention for Efficient Language Models
Paper • 2402.19427 • Published • 52 -
Beyond Language Models: Byte Models are Digital World Simulators
Paper • 2402.19155 • Published • 49 -
StarCoder 2 and The Stack v2: The Next Generation
Paper • 2402.19173 • Published • 130 -
Simple linear attention language models balance the recall-throughput tradeoff
Paper • 2402.18668 • Published • 18