Polynomial Context-Truncation Sensitivity

Polynomial Context-Truncation Sensitivity in Autoregressive Language Models: Sequential Wyner-Ziv Bounds for KV Cache Compression

arXiv:2605.25085v2 Announce Type: replace Abstract: We study the rate-distortion limits of online KV cache compression in autoregressive language models, formulating it as sequential Wyner-Ziv source coding on the filtration induced by the model, with the next-step query as decoder side information. Empirically, across four models spanning two families and $0.5$-$3$B parameters, we find that the next-token distribution's sensitivity to context truncation decays \emph{polynomially} rather...