Long-Context Language Modeling with Parallel Context Encoding

arXiv:2402.16617v1 Announce Type: new
Abstract: Extending large language models (LLMs) to process longer inputs is crucial for numerous applications. However, the considerable computational cost of transformers, coupled with limited generalization of positional encoding, restricts the size of their context window. We introduce Context Expansion with Parallel Encoding (CEPE), a framework that can be applied to any existing decoder-only LLMs to extend their context window. CEPE adopts a small encoder to process long inputs chunk by chunk and enables the frozen …

abstract applications arxiv computational context context window cost cs.cl encoding expansion framework inputs language language models large language large language models llms modeling positional encoding process transformers type