2025
PROMTEC: Fast LLM Inference Decoding using Prompt Multi-Lookup with Template Database and Common Sequences
ACL 2025finding
We propose PROMTEC, a novel multi-faceted approach to accelerate the inference of large language models (LLMs) by leveraging three key techniques: Prompt Multi-Lookup, Template Datastore, and Common Sequences methods. Prompt Multi-Lookup enhances the autoregressive decoding efficiency by generating…