Suchin Gururangan

@ssgrn

NLP research

Seattle, WA
kernelmachine.github.io
Vrijeme pridruživanja: studeni 2011.
16 fotografija ili videozapisa Fotografije i videozapisi

Tweetovi

Blokirali ste korisnika/cu @ssgrn

Jeste li sigurni da želite vidjeti te tweetove? Time nećete deblokirati korisnika/cu @ssgrn

  1. Prikvačeni tweet
    9. ruj 2019.

    In "Show Your Work", we propose a method to compare model performance in the context of computational budget. We also release a new AllenNLP hyperparameter search library! Check it out here:

    in "Show Your Work," we look at the status quo in experimental reporting in NLP -- it's abysmal -- and propose concrete ways to do better. to appear at EMNLP, by , , , , &
    3 proslijeđena tweeta 18 korisnika označava da im se sviđa
    Poništi
  2. proslijedio/la je Tweet
    21. sij

    AI2's MOSAIC team is looking for research interns! Come explore the frontiers of with us. The deadline for this round is Feb 5th – Please submit your application (and select MOSAIC as your "Desired Project Team") here:

    1 reply 19 proslijeđenih tweetova 46 korisnika označava da im se sviđa
    Poništi
  3. proslijedio/la je Tweet
    9. pro 2019.

    the kid on the bike claims sota, the one with the goggles is analyzing the model, and nobody is moving forward

    Tweet je nedostupan.
    27 proslijeđenih tweetova 201 korisnik označava da mu se sviđa
    Poništi
  4. proslijedio/la je Tweet
    5. stu 2019.

    When I learned that train_test_split is the most commonly used part of scikit-learn...

    20 replies 203 proslijeđena tweeta 1.252 korisnika označavaju da im se sviđa
    Poništi
  5. proslijedio/la je Tweet
    5. stu 2019.

    Show Your Work: Improved Reporting of Experimental Results. Paper by , , , and Paper URL:

    3 proslijeđena tweeta 10 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  6. proslijedio/la je Tweet
    5. stu 2019.

    Check out our poster on Quoref, a new reading comprehension dataset with questions requiring coreference resolution being presented at Session 11, 15:30 on Thursday, Nov. 7th, and chat with , , and about the work.

    5 proslijeđenih tweetova 32 korisnika označavaju da im se sviđa
    Poništi
  7. proslijedio/la je Tweet
    28. lis 2019.

    We () are looking to hire a senior research scientist! Please share broadly.

    26 proslijeđenih tweetova 44 korisnika označavaju da im se sviđa
    Poništi
  8. proslijedio/la je Tweet
    24. lis 2019.

    The team will consider predoctoral young investigator applications on November 15 and February 15! Apply here:

    32 proslijeđena tweeta 63 korisnika označavaju da im se sviđa
    Poništi
  9. proslijedio/la je Tweet
    24. lis 2019.

    Over the last few months, a group of recent NLP PhD applicants compiled some thoughts, perspectives, & advice on the application process. We're happy to share this blog post, and hope future applicants will find it helpful for years to come: [1/3]

    15 replies 124 proslijeđena tweeta 401 korisnik označava da mu se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  10. proslijedio/la je Tweet
    10. lis 2019.

    is hiring! We're looking for a talented senior engineer to help advance our open-source NLP research library. Join a dedicated team of researchers and engineers, work on all aspects of the project, and help direct its future. Learn more:

    11 proslijeđenih tweetova 10 korisnika označava da im se sviđa
    Poništi
  11. proslijedio/la je Tweet
    7. lis 2019.

    Have you built a model on top of AllenNLP? We're looking for examples to showcase. Send us yours:

    8 proslijeđenih tweetova 29 korisnika označava da im se sviđa
    Poništi
  12. 7. lis 2019.

    A project I've been helping out on! Hopefully it's useful for anyone wanting to speed up their pretraining/fine-tuning setups with -compatible LMs.

    At last, language model pretraining with PyTorch+TPUs Our code trains PyTorch BERT/RoBERTa on TPUs, which is faster and cheaper than GPUs. Also check the repo for a more detailed comparison between TPUs/GPUs on PyTorch/Tensorflow.
    Prikaži ovu nit
    6 proslijeđenih tweetova 35 korisnika označava da im se sviđa
    Poništi
  13. proslijedio/la je Tweet
    24. ruj 2019.

    gains for low-resource language NLP from polyglot language model pretraining ("Rosita"), and a new "decontextual probe" -- to appear at CoNLL, by Phoebe Mulcaire, , --

    1 reply 9 proslijeđenih tweetova 55 korisnika označava da im se sviđa
    Poništi
  14. proslijedio/la je Tweet
    16. ruj 2019.

    So this just happened: - a bot found a vulnerability in a dependency - a bot sent a PR to fix it - the CI verified the PR - a bot merged it - a bot celebrated the merge with a GIF

    260 replies 6.949 proslijeđenih tweetova 17.741 korisnik označava da mu se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  15. proslijedio/la je Tweet
    16. ruj 2019.

    Wired piece on "Show your work" (by , , , , , appearing at EMNLP 2019) and other reproducibility research:

    20 proslijeđenih tweetova 54 korisnika označavaju da im se sviđa
    Poništi
  16. proslijedio/la je Tweet
    13. ruj 2019.

    In starting a preliminary write-up on my most recent byte level language modeling work (enwik8) I've really been wishing for many of the metrics highlighted by 's recent "Show your work" paper.

    12 proslijeđenih tweetova 65 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  17. proslijedio/la je Tweet
    10. ruj 2019.

    I'm excited about this one! In "Show Your Work" (to appear at EMNLP) we advocate for better reporting of experimental results and propose budget-aware evaluation of models. by , , , , & (1/5)

    7 proslijeđenih tweetova 40 korisnika označava da im se sviđa
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  18. 6. ruj 2019.

    If you're at , come by the poster session at 5:30p (location #25) to discuss cheap, low-resource pretraining!

    10 korisnika označava da im se sviđa
    Poništi
  19. 22. kol 2019.

    /s/left-most column/right-most column

    Prikaži ovu nit
    Prikaži ovu nit
    Poništi
  20. 22. kol 2019.

    It's interesting to think about how and why these stereotypical structures emerge. Do they help with cross-modality transfer? Or are they some random consequence of evolution? /3

    1 reply
    Prikaži ovu nit
    Prikaži ovu nit
    Poništi

@ssgrn još nije objavio nijedan Tweet.

Čini se da učitavanje traje već neko vrijeme.

Twitter je možda preopterećen ili ima kratkotrajnih poteškoća u radu. Pokušajte ponovno ili potražite dodatne informacije u odjeljku Status Twittera.

    Možda bi vam se svidjelo i ovo:

    ·