[HN Gopher] 16-Bit to 1-Bit: Visual KV Cache Quantization for Ef...
___________________________________________________________________
16-Bit to 1-Bit: Visual KV Cache Quantization for Efficient
Multimodal LLMs
Author : PaulHoule
Score : 65 points
Date : 2025-03-05 16:09 UTC (4 days ago)
(HTM) web link (arxiv.org)
(TXT) w3m dump (arxiv.org)
| kadushka wrote:
| Have they published their code?
___________________________________________________________________
(page generated 2025-03-09 22:00 UTC)