Data Statements for NLP: a proposal to help address issues of exclusion, overgeneralization, underexposure, & reproducibility in NLP @emilymbender https://t.co/kWaDqa7Z64 pic.twitter.com/k6c56xHRVI
— Rachel Thomas (@math_rachel) June 6, 2019
Data Statements for NLP: a proposal to help address issues of exclusion, overgeneralization, underexposure, & reproducibility in NLP @emilymbender https://t.co/kWaDqa7Z64 pic.twitter.com/k6c56xHRVI
— Rachel Thomas (@math_rachel) June 6, 2019
If you want a sneek-peek in @YejinChoinka, @rown and co-workers work on GROVER (a 1.5 billion param GPT-2-like model), check this live tweet 👇
— Thomas Wolf (@Thom_Wolf) June 6, 2019
Interesting hints, results, and analysis!
Paper: https://t.co/aJJrrmQjJc
Demo: https://t.co/MDfzlYa1iE https://t.co/jjwQEUU0tI
Notes on the "Limitations of the Empirical Fisher Approximation" by Kunstner et al.: an excellently written discussion paper.https://t.co/GDTbzec3gt
— Ferenc Huszár🇪🇺 (@fhuszar) June 6, 2019
"The Unreasonable Effectiveness of Transformer Language Models in Grammatical Error Correction," Alikaniotis and Raheja, Grammarly: https://t.co/QsoxZqUJ2f
— Miles Brundage (@Miles_Brundage) June 6, 2019
For our fourth paper, we'll also leverage a Bayesian approach, but in the context of Bayesian optimization for multi-task learning:
— Sebastian Ruder (@seb_ruder) June 5, 2019
AUTOSEM: Automatic Task Selection and Mixing in Multi-Task Learninghttps://t.co/xmcLbBgaGZ
In a nice segue, we now move from words to subwords for our second paper:
— Sebastian Ruder (@seb_ruder) June 5, 2019
Subword-based Compact Reconstruction of Word Embeddings https://t.co/Nxav5QWqEa
First paper: How Large a Vocabulary Does Text Classification Need?
— Sebastian Ruder (@seb_ruder) June 5, 2019
A Variational Approach to Vocabulary Selection https://t.co/ng4R6bz9eU
Surprising how simple ideas can yield such a good generative model!
— Oriol Vinyals (@OriolVinyalsML) June 4, 2019
-Mean Squared Error loss on pixels
-Non-autoregressive image decoder
-Discrete latents w/ straight through estimator
w/ @catamorphist & @avdnoord
VQ-VAE-2: https://t.co/6ZKJDbcoTc
Code: https://t.co/KoaUvcLWQF pic.twitter.com/xhqB2v7Hk7
VQVAE-2 finally out!
— Aäron van den Oord (@avdnoord) June 4, 2019
Powerful autoregressive models in a hierarchical compressed latent space. No modes were collapsed in the creation of these samples ;)
Arixv: https://t.co/nQH5dzQhib
With @catamorphist and @vinyals
More samples and details 👇 [thread] pic.twitter.com/aIg6sk6aZt
Building a Language User Interface? Let Genie Generate It For You!
— Richard Socher (@RichardSocher) June 4, 2019
Blog: https://t.co/mIyBTaZAHZ
Paper: https://t.co/6u6SZF2LZm
One of the papers from my time as Stanford adj. prof last year. pic.twitter.com/5KjCXnjtg3
RF-Net: An End-to-End Image Matching Network based on Receptive Field. More work on CNNs for keypoint extraction and image matching. Seems like formulating this problem using deep learning is “easy” now. https://t.co/K2t3D0b3Kg #computervision pic.twitter.com/gXUSqtfpbD
— Tomasz Malisiewicz @ CVPR 2019 (@quantombone) June 4, 2019
Give AI the starting and ending parts of a video and it will generate likely in-between video sequences.
— DataScienceNigeria (@DataScienceNIG) June 3, 2019
An excellent work on inbetweening in stochastic video generation by @GoogleAI , which is generally approached by means of recurrent neural networks.https://t.co/MfnYXu9g4Z pic.twitter.com/PQZv1fVsf4