Zum Inhalt springen

Thomas Hikaru Clark

@thomashikaru.bsky.social did:plc:d4k5rp4iscij2fadtuwdczhs

researching language as a cognitive technology Faculty Fellow at NYU Center for Data Science 💬 🧠 💻 🗼 PhD @ MIT Brain&CogSci former HS math and CS teacher

Bei Bluesky seit

Follower
198
Folgt
162
Posts
16
Tage in Ranglisten
0

Zusammenfassung

Thomas Hikaru Clark (@thomashikaru.bsky.social) ist seit 10 April 2025 auf Bluesky. Das Konto hat 198 Follower, folgt 162 Konten und hat 16 Posts veröffentlicht.

Ranglisten-Platzierungen

Die besten Tage, an denen Posts dieses Kontos in Debyas Tagesliste kamen.

Dieses Konto war noch in keiner Tagesrangliste.

Zuletzt gelikte Posts

Öffentliche Likes aus dem Bluesky-Netzwerk.

  • Madeleine Daepp@mdaepp.bsky.social

    Thank you for this, I sure was feeling crazy! Lately the jargon has gotten so bad that I've had to go back to the old "explain this like I'm five" trick

    Post anzeigen
  • Rachel Ryskin@ryskin.bsky.social

    Noise models are also calibrated to experience. They adapt to context (readers expect more errors from an error-prone typist) and plausibly differ across populations, including older adults, people with aphasia, children, and L2 learners. 🧪 This opens up lots of new, testable predictions!

    Post anzeigen
  • Rachel Ryskin@ryskin.bsky.social

    The same inference can mislead us: “The mother gave the candle the daughter” is often read as if a “to” was dropped. This unifies explanations of phenomena from phonemic restoration to illusions (“No head injury is too trivial to be ignored”) to real-time processing effects (via noisy memory).

    Post anzeigen
  • Rachel Ryskin@ryskin.bsky.social

    We distinguish three sources of noise: errors in the speaker’s production (slips, typos), noise in the environment (a cough, a dropped call), and noise inside the comprehender (perception, attention, memory). Each has its own signature, and comprehenders seem to model all three.

    Post anzeigen
  • Rachel Ryskin@ryskin.bsky.social

    When you hear “Can you demonstrate what it’s like to brush our 𝘵𝘦𝘦𝘵𝘩’𝘴 𝘱𝘦𝘵?” you probably infer that “pet’s teeth” was intended. You get there by combining prior expectations about what people are likely to say with a model of how messages get corrupted along the way.

    Post anzeigen

Melde dich an, um mehr zu sehen

Blockierte Konten

Blockierungen sind auf Bluesky öffentlich.

Dieses Konto blockiert niemanden.