Kv Cache Steering Reasoning

Guide frozen language models toward multi-step reasoning by modifying cached key-value representations after the prefilling stage. Extract steering vectors from contrastive prompt pairs and apply them to KV cache with scalar coefficients. Improves reasoning on GSM8K, ARC, CommonsenseQA while adding only 10ms overhead per token.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit e96b1abefb

Frequently asked questions

npx get-skill add adu2021/kv-cache-steering-reasoning