Efficient Few-Shot Learning Without Prompts
— AK (@_akhaliq) September 23, 2022
abs: https://t.co/od6AtDNSJJ pic.twitter.com/C9uWHgOEXI
Efficient Few-Shot Learning Without Prompts
— AK (@_akhaliq) September 23, 2022
abs: https://t.co/od6AtDNSJJ pic.twitter.com/C9uWHgOEXI
Reading through OpenAI Whisper paper https://t.co/3PmWvQNCFs some notes: pic.twitter.com/QVeqaGVvsV
— Andrej Karpathy (@karpathy) September 22, 2022
I’m happy to share the published version of our ConVIRT algorithm, appearing in #MLHC2022 (PMLR 182). In 2020, this was a pioneering work in contrastive learning of perception by using naturally occurring paired text. Unfortunately, things took a winding path from there. 🧵👇 pic.twitter.com/CUwAZftKlV
— Christopher Manning (@chrmanning) September 21, 2022
We've trained a neural net called Whisper that approaches human-level robustness and accuracy on English speech recognition. It performs well even on diverse accents and technical language. Whisper is open source for all to use. https://t.co/ueVywYPEkK
— OpenAI (@OpenAI) September 21, 2022
Text2Light: Zero-Shot Text-Driven HDR Panorama Generation
— AK (@_akhaliq) September 21, 2022
abs: https://t.co/cwTRtf3AJF
project page: https://t.co/mDRWKQzA7r
github: https://t.co/6jJuhy37aT pic.twitter.com/qMhy5f75zE
LAVIS: A Library for Language-Vision Intelligence
— AK (@_akhaliq) September 20, 2022
abs: https://t.co/hPQTx9Wu8L
github: https://t.co/LeHK7zxqDw pic.twitter.com/6s1x7D80H1
Introducing PaLI, a new jointly-scaled multilingual language-image model that's built on 10B images and tens of billions of alt-texts and OCR annotations, in (wait for it!) over 100 languages. #pali #pathways Learn more and read the paper → https://t.co/5KEhtG7P65 pic.twitter.com/9vLmrecoCM
— Google AI (@GoogleAI) September 15, 2022
CLIP-ViP: Adapting Pre-trained Image-Text Model to Video-Language Representation Alignment
— AK (@_akhaliq) September 15, 2022
abs: https://t.co/Wjx8nicsLJ pic.twitter.com/qLNPpomFdR
Text-Free Learning of a Natural Language Interface for Pretrained Face Generators
— AK (@_akhaliq) September 9, 2022
abs: https://t.co/MdUAsWgJ8e pic.twitter.com/NnH4V5D9Qf
colab: https://t.co/VDRk57H3QU
— AK (@_akhaliq) September 4, 2022
COYO-700M, a large-scale dataset that contains 747M image-text pairs as well as many other meta-attributes to increase the usability to train various models
— AK (@_akhaliq) September 2, 2022
github: https://t.co/3J3hh5ocSu pic.twitter.com/nTbN9MztnD
Faithful Reasoning Using Large Language Models
— AK (@_akhaliq) August 31, 2022
abs: https://t.co/PKQDgzCdF3 pic.twitter.com/Mia8kRS8iG