InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory
Paper
•
2402.04617
•
Published
•
6
Long-context
-
-
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 • 23 -
Same Task, More Tokens: the Impact of Input Length on the Reasoning Performance of Large Language Models
Paper • 2402.14848 • Published • 19 -
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
Paper • 2402.13753 • Published • 116 -
LongAgent: Scaling Language Models to 128k Context through Multi-Agent Collaboration
Paper • 2402.11550 • Published • 19 -
In Search of Needles in a 10M Haystack: Recurrent Memory Finds What LLMs Miss
Paper • 2402.10790 • Published • 42 -
Make Your LLM Fully Utilize the Context
Paper • 2404.16811 • Published • 55