Katherine Tian

icon indicating a website link https://kttian.github.io/

icon location Bay Area, CA icon location math and cs @ harvard, research in generative ml / accurate text generation

Results 1 comments of Katherine Tian

Blip2 vicuna instruct

That could work - what is the 'text_output' field intended to represent? But also, I mostly want to replicate the authors' training first! None of the datasets (e.g. okvqa) currently...