EFIM rearranges the fill-in-the-middle prompt so prefix and suffix stay fixed across user edits, enabling more KV cache reuse, while a fragment tokenization training method fixes the resulting partial-word generation problem.
arXiv (2021)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
EFIM: Efficient Serving of LLMs for Infilling Tasks with Improved KV Cache Reuse
EFIM rearranges the fill-in-the-middle prompt so prefix and suffix stay fixed across user edits, enabling more KV cache reuse, while a fragment tokenization training method fixes the resulting partial-word generation problem.