Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
vov_or
on Feb 28, 2023
|
parent
|
context
|
favorite
| on:
Beating OpenAI CLIP with 100x less data and comput...
Yes, it is possible. Approaches, on which our model is based, are capable to solve VQA and other similar tasks showing SOTA results.
ilaksh
on Feb 28, 2023
|
next
[–]
Do you know anyone working on a large text completion model based on it?
freediver
on Feb 28, 2023
|
prev
[–]
Do you have/plan to have a text embeddings model?
vov_or
on Feb 28, 2023
|
parent
[–]
Yes, we are training text embedding models right now. And also have plans to open-source some of them! In addition, we train encoders for different modalities with retrieval purposes. For example, video data.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
