Valency Hub
alphaFeedback

You’re reading v1 — jump to latest

WorksComputer Science

Prefix Keys and Values Matter Late: RIPE Caching for Self-Speculative Decoding in Looped Language Models

Dezhen Lu, Runguo Li, Zixiang Ni

Published Fresh — added to the Hub within the last 30 daysv1

Publication details
License
arXiv nonexclusive license
Status
Published
AI contributions
Not recorded for this paper
View details

Versions

Each version keeps its own PDF, citation, and review record. Endorsements stay with the version they examined.

2 versionsNewest first

  1. v2

    LatestRead version

  2. v1

    Read version
    Viewing

Beginning of version history