Based on a true story

Decoupling Reasoning from Observations for Efficient Augmented Language Models

[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.

Materials for the Learn PyTorch for Deep Learning: Zero to Mastery course.