Collection by 
Apr 29, 2024
-
InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory
Paper • 2402.04617 • Published • 6 -
BurstAttention: An Efficient Distributed Attention Framework for Extremely Long Sequences
Paper • 2403.09347 • Published • 22 -
Resonance RoPE: Improving Context Length Generalization of Large Language Models
Paper • 2403.00071 • Published • 24 -
Training-Free Long-Context Scaling of Large Language Models
Paper • 2402.17463 • Published • 24
