FewNLU: Benchmarking State-of-the-Art Methods
— AK (@ak92501) September 28, 2021
for Few-Shot Natural Language Understanding
pdf: https://t.co/IhdNEnGWo8
abs: https://t.co/YO2hqCTWnk pic.twitter.com/eF7icGRVbA
FewNLU: Benchmarking State-of-the-Art Methods
— AK (@ak92501) September 28, 2021
for Few-Shot Natural Language Understanding
pdf: https://t.co/IhdNEnGWo8
abs: https://t.co/YO2hqCTWnk pic.twitter.com/eF7icGRVbA
"torch.manual seed(3407) is all you need: On the influence of random seeds in deep learning architectures for computer vision" https://t.co/LoQhOzpbVw. Results are actually not as bad as the title makes it seem. But yeah, reporting std dev or CIs should be(come) the default. pic.twitter.com/K2JElhgwCs
— Sebastian Raschka (@rasbt) September 27, 2021
Transformers Generalize Linearly
— AK (@ak92501) September 27, 2021
abs: https://t.co/ud0iUEYDyx
Transformers fail to generalize hierarchically across a wide variety of grammatical mapping tasks, but they exhibit an even stronger preference for linear generalization than comparable recurrent networks pic.twitter.com/VzbM2SQTZl
Muzic: Music Understanding and Generation with Artificial Intelligence
— AK (@ak92501) September 25, 2021
github: https://t.co/2XtiUh3h8e pic.twitter.com/4PTPX4qMia
Amusing! Object detection cast naively into language modeling framework + borrowing many of the tips&tricks.
— Andrej Karpathy (@karpathy) September 24, 2021
- random object ordering seems fine ✅
- coords, class labels flattened into a single softmax 😂
- sequence augmentation is the most gnarly part, almost as yucky as nms 😬 https://t.co/FxSz5UbpxY
Have you wondered why object detection, unlike classification, has so many sophisticated algorithms?
— Ting Chen (@tingchenai) September 23, 2021
With Pix2Seq (https://t.co/ygsG3aAIbG), we simply cast object detection as a language modeling task conditioned on pixels!
(with @srbhsxn, Lala Li, @fleet_dj, @geoffreyhinton) pic.twitter.com/aTYZ5IvJc9
Neural Distance Embeddings for Biological Sequences
— AK (@ak92501) September 22, 2021
abs: https://t.co/fvlN4fcGpB
github: https://t.co/Z0Ok68cWhH pic.twitter.com/nmNGqO8IRy
Primer: Searching for Efficient Transformers for Language Modeling
— AK (@ak92501) September 20, 2021
abs: https://t.co/JM9v7pNoSI
github: https://t.co/xhA7uGyC7H
Experiments show Primer’s gains over Transformer increase as compute scale grows and follow a power law with respect to quality at optimal model sizes pic.twitter.com/CXq1yYMfUA
Wish your neural networks faster and more accurate?
— Mingxing Tan (@tanmingxing) September 16, 2021
Check out our recent EfficientNetV2 and CoAtNet, which significantly speed up the training and inference, while achieving state-of-the-art 90.88% top-1 accuracy on ImageNet. https://t.co/9buCSZmYby
An Empirical Study of GPT-3 for Few-Shot Knowledge-Based VQA
— AK (@ak92501) September 13, 2021
abs: https://t.co/A3s6zdW0Iu
a simple yet effective method that Prompts GPT3 via the use of Image Captions. Using only 16 examples, PICa surpasses the supervised sota by an absolute +8.6 points on the OK-VQA dataset pic.twitter.com/9HKybsk1qu
ConvMLP: Hierarchical Convolutional MLPs for Vision
— AK (@ak92501) September 10, 2021
pdf: https://t.co/f6c1XmyLSX
abs: https://t.co/vkgXvlCmcD
github: https://t.co/FyjNR8W3Oq pic.twitter.com/RXXJgzoula
Efficient Nearest Neighbor Language Models
— AK (@ak92501) September 10, 2021
pdf: https://t.co/uEtibkYA1L
abs: https://t.co/K0hBVhtvlk pic.twitter.com/ZKd0vnBDCs