VisHead-Spec: Head-Routed Visual Attention for Multimodal Speculative Decoding
arXiv:2611.21839v1 Announce Type: new Abstract: Speculative decoding accelerates vision-language models(VLMs) while preserving the target model's outputs, but its speedup diminishes under high concurrency serving as the system becomes more compute bounded as the batch size increases. This problem is pronounced for multimodal speculative decoding as the long range of the image tokens in the prompt can significantly increase the computational cost during draft token prediction. The block-diffusion drafters (e.g. DFlash, DSpark) can effectively speed up the draft model efficiency by predicting mu
阅读原文 · arXiv 机器学习本文由 SoccerDream AI 自动聚合索引,内容版权归原作者与 arXiv 机器学习 所有。
本站仅作技术动态聚合,不构成任何形式的建议。
本站仅作技术动态聚合,不构成任何形式的建议。