‹ BackHN Continuity

Thread

LensVLM: Compressing long context as images, expanding only relevant pages

91 points · 10 comments · victormustar

  1. taylorfinley · · focus · HN ↗
    Oh My Pi has done this for a while now, they call it Snap compact.
    1. 2muchtime · · focus · HN ↗
      Ha! Didn’t realize that’s what it was doing, I’d compact and it would say snap compact with a little icon of a camera, so this all makes sense now.
    2. dvt · · focus · HN ↗
      I remember reading a paper entitled "A Picture is Worth a Thousand Tokens" or something similar like 2-3 years ago. The reality is that no one really wants/needs contexts that big, anyway. It's hard enough making LLMs truly useful even with a small/medium context.
      1. wangii · · focus · HN ↗
        yep, deepseek
Open on Hacker News to reply ↗

Unofficial Hacker News client; not affiliated with Y Combinator.