-
Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper • 2404.08801 • Published • 62 -
Ring Attention with Blockwise Transformers for Near-Infinite Context
Paper • 2310.01889 • Published • 8 -
World Model on Million-Length Video And Language With RingAttention
Paper • 2402.08268 • Published • 33 -
Scaling Transformer to 1M tokens and beyond with RMT
Paper • 2304.11062 • Published • 2
Collections
Discover the best community collections!
Collections including paper arxiv:2309.00071
-
Sequence Parallelism: Long Sequence Training from System Perspective
Paper • 2105.13120 • Published • 5 -
Ring Attention with Blockwise Transformers for Near-Infinite Context
Paper • 2310.01889 • Published • 8 -
Striped Attention: Faster Ring Attention for Causal Transformers
Paper • 2311.09431 • Published • 4 -
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
Paper • 2309.14509 • Published • 16
-
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
Paper • 2402.13753 • Published • 106 -
Data Engineering for Scaling Language Models to 128K Context
Paper • 2402.10171 • Published • 18 -
LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
Paper • 2402.11550 • Published • 12 -
The What, Why, and How of Context Length Extension Techniques in Large Language Models -- A Detailed Survey
Paper • 2401.07872 • Published • 2
-
Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Paper • 2312.00752 • Published • 131 -
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 59 -
Soaring from 4K to 400K: Extending LLM's Context with Activation Beacon
Paper • 2401.03462 • Published • 25 -
Extending LLMs' Context Window with 100 Samples
Paper • 2401.07004 • Published • 14
-
Soaring from 4K to 400K: Extending LLM's Context with Activation Beacon
Paper • 2401.03462 • Published • 25 -
MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformers
Paper • 2305.07185 • Published • 8 -
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 59 -
Infinite-LLM: Efficient LLM Service for Long Context with DistAttention and Distributed KVCache
Paper • 2401.02669 • Published • 11
-
LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models
Paper • 2309.12307 • Published • 82 -
NEFTune: Noisy Embeddings Improve Instruction Finetuning
Paper • 2310.05914 • Published • 13 -
SOLAR 10.7B: Scaling Large Language Models with Simple yet Effective Depth Up-Scaling
Paper • 2312.15166 • Published • 55 -
Soaring from 4K to 400K: Extending LLM's Context with Activation Beacon
Paper • 2401.03462 • Published • 25
-
Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation
Paper • 2108.12409 • Published • 4 -
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 59 -
MIMIC-IT: Multi-Modal In-Context Instruction Tuning
Paper • 2306.05425 • Published • 9 -
Music ControlNet: Multiple Time-varying Controls for Music Generation
Paper • 2311.07069 • Published • 43