Co-occurring keywords · papers mentioning both
Papers
Length Generalization of Causal Transformers without Position Encoding
ACL 2024
LM-Infinite: Zero-Shot Extreme Length Generalization for Large Language Models
NAACL 2024
LongHeads: Multi-Head Attention is Secretly a Long Context Processor
EMNLP 2024