Diffusionvl Ar To Diffusion

Convert pre-trained autoregressive vision-language models into diffusion VLMs without architectural modifications. Use block diffusion strategy enabling arbitrary-length generation and KV-cache reuse. Hybrid attention enforces bidirectional within blocks, causal between blocks. Requires less than 5% of data compared to prior diffusion VLM methods.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 715a52662e

Frequently asked questions

npx get-skill add adu2021/diffusionvl-ar-to-diffusion