GPT-3 has 175 billion parameters, trained on 300 billion tokenshttps://t.co/rE97CQclwl pic.twitter.com/5tJgwwmABN
— Mark Riedl (@mark_riedl) May 29, 2020
GPT-3 has 175 billion parameters, trained on 300 billion tokenshttps://t.co/rE97CQclwl pic.twitter.com/5tJgwwmABN
— Mark Riedl (@mark_riedl) May 29, 2020
New Video 🥳🎉Facebook's DETR Object Detection Explained!https://t.co/ef51gqllNR@alcinos26 @fvsmassa @syhw @szagoruyko5 @facebookai
— Yannic Kilcher (@ykilcher) May 28, 2020
AutoSweep: Recovering 3D Editable Objects
— roadrunner01 (@ak92501) May 28, 2020
from a Single Photograph
pdf: https://t.co/TelxiBAGw6
abs: https://t.co/AvaNH12Tv3
webpage: https://t.co/jVmOvi7nMB pic.twitter.com/b4W67ocszq
- Take FasterRCNN
— Soumith Chintala (@soumithchintala) May 27, 2020
- Remove clunky NMS, Proposals, ROIAlign, Refinement and their gazillion hyperparameters
- Replace with Transformer
- Win!
Simplifies code and improves performance.
Nice work from @fvsmassa (torchvision maintainer) and his collaborators at FAIR. https://t.co/QeIq6IluhN
Learning To Classify Images Without Labels
— roadrunner01 (@ak92501) May 27, 2020
pdf: https://t.co/VkrV9FUOaN
abs: https://t.co/8PIwSb7857 pic.twitter.com/W4FIHB3dBJ
Thrilled to share new work! “Retrieval-Augmented Generation for Knowledge-Intensive NLP tasks”.
— Patrick Lewis (@PSH_Lewis) May 26, 2020
Big gains on Open-Domain QA, with new State-of-the-Art results on NaturalQuestions, CuratedTrec and WebQuestions.
check out here: https://t.co/SVZ6K4tDn5.
1/N pic.twitter.com/w4CwLxiWxr
An implementation of World Models using TensorFlow 2.2 that reproduces all the experiments. Includes Jupyter notebook for visualization. https://t.co/WaSnMDHTdI pic.twitter.com/7j5jOadDpY
— hardmaru (@hardmaru) May 26, 2020
Heads up- I've been waiting for months to see this happen and it finally did: BiT-M repo/pretrained models are up, and they're even providing PyTorch versions! https://t.co/Zr4GACAlcR
— Jason #Masks4All Antic (@citnaj) May 24, 2020
Wish You Were Here: Context-Aware Human Generation
— roadrunner01 (@ak92501) May 22, 2020
pdf: https://t.co/h8VWmwt5Xw
abs: https://t.co/BUAml4pJWE pic.twitter.com/uQRuJwVH3q
Latent Adversarial Generator Code is Out!
— David Berthelot (@D_Berthelot_ML) May 21, 2020
Code: https://t.co/2XmNyIcEVv
Arxiv: https://t.co/a8HqVpAl2r@docmilanfar @goodfellow_ian pic.twitter.com/O64yKLJMQ7
Very cool research from @Waymo / @GoogleAI on using graph attention for multi-agent vehicle motion forecasting!https://t.co/2t7l0JJWIu pic.twitter.com/PpiDqvNy7Z
— Petar Veličković (@PetarV_93) May 21, 2020
FashionBERT: Text and Image Matching with Adaptive Loss
— roadrunner01 (@ak92501) May 21, 2020
for Cross-modal Retrieval
pdf: https://t.co/jW3Rcmta56
abs: https://t.co/Ck6XtdM0AH pic.twitter.com/AiTKxPD59L