vov_or
Joined 4 years ago
Karma
285
Submissions
28
Member since
Mar 23
Senior NLP research engineer
View on Hacker NewsRecent submissions
Comment · 3 years ago
Guys trained a multi-modal chatbot with visual and language instructions based on the open-source multi-modal model OpenFlamingo!
Paper link: https://arxiv.org/abs/2305.04790
Comment · 3 years ago
Hi!
MSCOCO and Flickr datasets are the main datasets for Image retrieval. The results published in most papers (including CLIP) are based on them. So we used exactly these datasets for evaluation.
Comment · 3 years ago
Hi!
You are right that we had to clarify that "100 times better at retrieval".
Btw, we have plans to tune models, evaluate, and publish results in different tasks (zero-shot ImageNet classification, etc)
Comment · 3 years ago
More efficient - for sure!