This interview includes many reasons for why we are building https://t.co/2tfrBwhwOc, what's wrong with Google, how you can do better than Google (->AI, apps, privacy, and personalization)https://t.co/njACJcnw8R
— Richard Socher (@RichardSocher) June 22, 2022
I really enjoyed talking to @linked_do and his insightful questions
Square One Bias in NLP: Towards a Multi-Dimensional Exploration of the Research Manifold (ACL 2022)
— Sebastian Ruder (@seb_ruder) June 22, 2022
We identify the current prototypical NLP experiment (the “square one”) and assess the dimensions along which NLP papers go beyond this prototype.https://t.co/EtWN1sOz3Z pic.twitter.com/gW4XxpuScQ
I signed on this because I am generally against deployment of AI for harmful purposes, even in jest. However, the Stanford folks pitchforking a controversy as a marketing opportunity to advance “foundation models” terminology, and their newly created “center” is just bad taste. https://t.co/Bm70coyAki
— Delip Rao (@deliprao) June 22, 2022
Global Context Vision Transformers
— AK (@_akhaliq) June 22, 2022
abs: https://t.co/d6go0yv7fu
github: https://t.co/rUYFs09ReC
On ImageNet-1K dataset for classification, the base, small and tiny variants of GC ViT with 28M, 51M and 90M parameters achieve 83.2%, 83.9% and 84.4% Top-1 accuracy, respectively pic.twitter.com/XKoJAvUcYm
Temporally Consistent Semantic Video Editing
— AK (@_akhaliq) June 22, 2022
abs: https://t.co/sg1dRt2xkw
project page: https://t.co/PyZKnxUQko pic.twitter.com/1Az9nG5ccH
DALL-E for Detection: Language-driven Context Image Synthesis for Object Detection
— AK (@_akhaliq) June 22, 2022
abs: https://t.co/rXx4npbY5G pic.twitter.com/QBHP494eSn
iBoot: Image-bootstrapped Self-Supervised Video Representation Learning
— AK (@_akhaliq) June 18, 2022
abs: https://t.co/dkZUd4QC81 pic.twitter.com/pJFpxd7ckU
Disentangling visual and written concepts in CLIP
— AK (@_akhaliq) June 18, 2022
abs: https://t.co/VsyuDV4HNI
project page: https://t.co/2hTQnhR2o1 pic.twitter.com/LbWpnpTTHT
Deep nets can be overconfident (and wrong) on unfamiliar inputs. What if we directly teach them to be less confident? The idea in RCAD ("Adversarial Unlearning") is to generate images that are hard, and teach it to be uncertain on them: https://t.co/lJf5aVv3Jr
— Sergey Levine (@svlevine) June 17, 2022
A thread: pic.twitter.com/pHj76WHnUA
Action recognition is a challenging task but has seen improvements through multimodality. In this latest blog post, learn how @Disney uses PyTorch to improve activity recognition through multimodal approaches. https://t.co/Xqy79N7Eh9 pic.twitter.com/t5kJq5msxx
— PyTorch (@PyTorch) June 17, 2022
Beyond Supervised vs. Unsupervised: Representative Benchmarking and Analysis of Image Representation Learning
— AK (@_akhaliq) June 17, 2022
abs: https://t.co/SunKnQH3NJ
project page: https://t.co/MgIrCZlQJv pic.twitter.com/yMHmZmk3J2
GoodBye WaveNet -- A Language Model for Raw Audio with Context of 1/2 Million Samples
— AK (@_akhaliq) June 17, 2022
abs: https://t.co/XRTTRbABXG
propose a generative auto-regressive architecture that can model audio waveforms over quite a large context, greater than 500,000 samples pic.twitter.com/2fJ9EUYb4u
