Tweetovi

Blokirali ste korisnika/cu @yoavgo

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

  1. Prikvačeni tweet
    6. ruj 2016.

    everyone is about Firth 1957 (you shall know a word...). somehow we all skipped Firth 1935!

    Poništi
  2. proslijedio/la je Tweet
    3. velj

    Check out BREAK - a new NLU benchmark for testing the ability of models to break down a question into the required steps for computing its answer. A work by Tomer Wolfson, accepted to TACL 2020.

    Poništi
  3. proslijedio/la je Tweet
    3. velj

    New TACL paper involving a lot of hard work from my twitter-less student Tomer, along with great collab. at AI2 and TAU. Paper/website at 1/2

    Prikaži ovu nit
    Poništi
  4. 2. velj

    not all my slides have muppet pictures. but p(slide being shared | muppet) is really high!

    Poništi
  5. 1. velj

    כותבים מבחנים.

    Poništi
  6. 24. sij

    MI, MIP, MIPP, MIPPP, MIPPPP, MIPPPPP, ...

    Poništi
  7. 22. sij

    זה קצת עצוב שהמעצב באתר של סטימצקי בחר להציג את כל הספרים כחפצים תלת מימדיים, אבל כמעט בלי נפח. וגם קצת מצחיק. וממש לא עושה חשק לקנות.

    Poništi
  8. 20. sij

    (for those who don't know the term, i mean these rectangles-instead-of-text thingies. )

    Prikaži ovu nit
    Poništi
  9. 20. sij

    ux rant: how did skeleton components in list/feed content become a thing. they are the most annoying thing ever. just show stuff as they are ready, without the placeholder. (the original skeletons were kinda legit, but this current version is just, ugh.)

    Prikaži ovu nit
    Poništi
  10. 16. sij

    והיום למדתי שזה חלק מפרוייקט של המרכז לערים חכמות בבר אילן, ושחוץ מהצמחיה יש שם המון חיישנים שמודדים דברים, כדי לבחון את ההשפעה של הצמחיה על הבניין והסביבה. מגניב!

    Prikaži ovu nit
    Poništi
  11. 16. sij

    לפני כמה חודשים ציפו את קירות אחד הבניינים היותר מכוערים בקמפוס במלא עדניות ירוקות. חשבתי לעצמי ״או! יפה שדואגים ככה לאסטטיקה״, ואז ״אוי אבל למה רק לבניין הזה ולא לשכן המסכן שלו״,

    Prikaži ovu nit
    Poništi
  12. proslijedio/la je Tweet
    14. sij

    Happy to release NN4NLP-concepts! It's a typology of important concepts that you should know to implement SOTA NLP models using neural nets: 1/3 We'll reference this in CMU CS11-747 this year, trying to maximize coverage. 1/3

    Prikaži ovu nit
    Poništi
  13. 13. sij

    but these are rare, and, lets face it, rather thank-less efforts for most of the people involved.

    Prikaži ovu nit
    Poništi
  14. 13. sij

    one project to applaud here is the Universal Dependencies, where there really is a lot of thought and attention and discussion and efforts going into getting the details just right. also, various workshop competitions in bionlp etc, where the teams have a real stake at the task.

    Prikaži ovu nit
    Poništi
  15. 13. sij

    at the same time, there is practically zero incentive in academia to get the details right. it is just "engineering" and "post processing" and "using knowledge" and "cleaning the data". where is the innovation?? but really what we need is getting the details right.

    Prikaži ovu nit
    Poništi
  16. 13. sij

    a HUGE problems in NLP, imo, is that to make actual progress, or to build something which is actually good, you just cannot escape doing very elaborate and laborious work on the messy details. language is messy, and we cannot esccape that.

    Prikaži ovu nit
    Poništi
  17. 8. sij
    Prikaži ovu nit
    Poništi
  18. 8. sij

    ובכן מה שקרה הבוקר זה שבעודי עומד ברחוב מחכה לאוטובוס תוך שיחת סקייפ תלת יבשתית, פתאום נפלה חתיכה די גדולה של חיפוי קיר מהבניין ליד, התנפצה על הרצפה ברעש גדול, מפספסת בכשתי שניות מישהי שעברה שם. פחד אלוהים. אז א תזהרו מבניינים, ו-ב, למי מדווחים על זה? יש שם עוד חיפוי שלא נפל.

    Prikaži ovu nit
    Poništi
  19. proslijedio/la je Tweet
    6. sij

    How do you pronounce "data"? RT for wider sample, comment for other

    Prikaži ovu nit
    Poništi
  20. 6. sij

    איבוד שפה טבעית.

    Prikaži ovu nit
    Poništi
  21. 6. sij
    Prikaži ovu nit
    Poništi

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:

    ·