Get the latest tech news

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


We’re on a journey to advance and democratize artificial intelligence through open source and open science.

None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of images

images

Photo of long context

long context

Photo of relevant pages

relevant pages

Related news:

News photo

Interpol uses AI to identify 126 terrorists by analyzing over 100,000 images with new facial recognition tools

News photo

Hackers find 1.6 million images on a single Flock camera, with easily beaten encryption

News photo

Boomers love AI images of their grandkids; millennial parents hate it