Xiangyu Hong's starred repositories
ChunkLlama
[ICML'24] Data and code for our paper "Training-Free Long-Context Scaling of Large Language Models"
attention_with_linear_biases
Code for the ALiBi method for transformer language models (ICLR 2022)
landmark-attention
Landmark Attention: Random-Access Infinite Context Length for Transformers
streaming-llm
[ICLR 2024] Efficient Streaming Language Models with Attention Sinks