works on
inference optimization 1large language models 1latent attention 1model conversion 1speculative decoding 1

From the 1 of 1 linked paper with an AI index.

collaborators

1 paper

cs.LG2026

Beyond KV Reconstruction: Functional Reconstruction for MLA Draft Models in Speculative Decoding

Weiye Shi, Fanxu Meng, Muhan Zhang

The paper proposes a functional reconstruction method to convert multi-head or grouped-query attention models into multi-head latent attention form, preserving cache efficiency and…