Idan Blank

@IdanAsherBlank

Cognitive (neuro)scientist studying language at UCLA Psych (BlankLangLab). Views are my own. He/Him. RT ≠ endorsement. 🏳️‍🌈

Vrijeme pridruživanja: siječanj 2016.

Tweetovi

Blokirali ste korisnika/cu @IdanAsherBlank

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

  1. 2. velj

    Good news to all the grad students who have been wasting years doing backflips trying to make some priming / IQ training / bilingual advantage / etc work - it's not our fault, you're just incompetent.

    Poništi
  2. proslijedio/la je Tweet
    1. velj

    "These results provide no evidence for a replication crisis in cognitive psychology. It is therefore unwarranted to generalize from attention-grabbing replication failures in social psychology to cognitive psychology."

    Poništi
  3. 2. velj

    Where is "forced to sit through a 5h student production of Hamlet"?

    Poništi
  4. proslijedio/la je Tweet
    2. velj

    Causes of death in Shakespeare plays

    Poništi
  5. proslijedio/la je Tweet
    2. velj

    It's 02.02.2020, the day for sharing palindromes, and this one's a doozy. "Doc, note: I dissent. A fast never prevents a fatness. I diet on cod."

    Poništi
  6. 1. velj

    Linguistic prediction, both surface-level/linear and hierarchical, relies on language-specific resources, not domain-general, top-down mechanisms. Naturalistic fMRI and a new analytical method. Now out in Neuropsychologia; 50 days free access here:

    Poništi
  7. 1. velj

    Just said to our applicants: “meet theccurrent students near the fountain and they’ll ride you. Uhh... they’ll give you a ride.”

    Poništi
  8. 31. sij

    I was terrified by their work in 2010, and this just reinforces my fears that if climate change won't kill us - this will. Terror and awe.

    Poništi
  9. proslijedio/la je Tweet
    30. sij

    Boston Dynamics' Atlas shows off its agility with a flawless gymnastic routine. Credit: Boston Dynamics

    Poništi
  10. 30. sij

    Apologies for how crass this is, BUT: creative use of the “way” construction!

    Poništi
  11. proslijedio/la je Tweet
    29. sij

    My lab, the Stanford Language and Cognition Lab, is looking for a new lab manager! We study language learning and cognitive development. Lots of ways to get involved in research and learn new skills/methods in a fun, supportive environment. Come join us!

    Poništi
  12. 28. sij

    I also learned I need to pay almost $4000 out of pocket to cover the applicants’ hotel rooms b/c the department can only reimburse, not pay directly. Of course the youngest faculty would be best situated to pay this amount.

    Prikaži ovu nit
    Poništi
  13. 28. sij

    Department: we don’t assign assistant professors many service duties; that’s for post-tenure! Also department: our tradition is that the youngest faculty organizes the interview weekend for our grad program omg that’s you lol 🤷🏻‍♂️

    Prikaži ovu nit
    Poništi
  14. 27. sij

    By far, the biggest price I pay compared to my native-speaking friends isn’t writing (e.g., some of them obsess over crafting emails, too), but how long it takes me to read—student essays for grading, grad school apps, *papers*... and it’s mentally exhausting.

    Poništi
  15. proslijedio/la je Tweet
    26. sij

    Who here realises the ENORMOUS privilege that the use of English as an international science & publishing language conjures on native English speakers in academia?

    Prikaži ovu nit
    Poništi
  16. 27. sij

    Yes, but you don't write "people are claiming GPT-2 is intelligent and this conflates thought & language"; you write "GPT-2's failures that I demonstrate here tell us something about nature, nurture, and the human mind". They really don't.

    Poništi
  17. 26. sij

    ... either by computational models learning the correct inferences from input (even realistic, e.g., sparse & noisy, input), or by showing the relevant structures serve a communicative function (non-communicativeness being the original reason for assuming unlearnability). 6/6

    Prikaži ovu nit
    Poništi
  18. 26. sij

    p.s.2 I personally think we’re born with important inductive biases related to intuitive theories about how the world works. But these are *not* language, or about language. So far, most putative innate language-specific biases have been undermined... 5/6

    Prikaži ovu nit
    Poništi
  19. 26. sij

    p.s.1 There’s still ongoing work about *whether* vectors learned by such networks are indeed not (implicitly) symbolic (e.g., Smolensky). So whatever their failures (the ones relevant to language, not the Marcus ones), we don’t know for sure they’re due to lack of symbols. 4/6

    Prikaži ovu nit
    Poništi
  20. 26. sij

    3. Even much simpler networks appear (to my unending surprise and “disbelief”) to learn a lot about language itself (e.g., many, but not all, phenomena related to long-distance dependencies). 3/6

    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:

    ·