Skip to content
Hallway
0:00 / 0:00

#psycholinguistics#llms#syntax

Edward Gibson: Human Language, Psycholinguistics, Syntax, Grammar & LLMs | Lex Fridman Podcast #426

Lex Fridman

5M subscribers • 39 videos on Hallway

YouTube
  • Recommended by 2 people: 2 on Hacker News.
  • Presented by Edward Gibson, professor of psycholinguistics at MIT

Intermediate. Assumes working knowledge.

471K views2 years ago2 recommendations

AI and machine learningComputer scienceTransformers and LLMsComputation and complexity

Edward Gibson discusses the relationship between human language, psycholinguistics, syntax, grammar, and how large language models process language.

Edward Gibson is a psycholinguistics professor at MIT and heads the MIT Language Lab. Please support this podcast by checking out our sponsors: - Yahoo Finance: yahoofinance.com - Listening: listening.com/lex and use code LEX to get one month free - Policygenius: policygenius.com/lex - Shopify: shopify.com/lex to get $1 per month trial - Eight Sleep: eightsleep.com/lex to get special savings TRANSCRIPT: lexfridman.com/edward-gibson-transcript EPISODE LINKS: Edward's X: x.com/LanguageMIT TedLab: tedlab.mit.edu/ Edward's Google Scholar: scholar.google.com/citations?user=4FsWE64AAAAJ TedLab's YouTube: youtube.com/@Tedlab-MIT PODCAST INFO: Podcast website: lexfridman.com/podcast Apple Podcasts: apple.co/2lwqZIr Spotify: open.spotify.com/episode/0VGHDfaIxJsiG48bokSxz7 RSS: lexfridman.com/feed/podcast/ Full episodes playlist: youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4 Clips playlist: youtube.com/playlist?list=PLrAXtmErZgOeciFP3CBCIEElOJeitOr41 OUTLINE: - Introduction - Human language - Generalizations in language - Dependency grammar - Morphology - Evolution of languages - Noam Chomsky - Thinking and language - LLMs - Center embedding - Learning a new language - Nature vs nurture - Culture and language - Universal language - Language translation - Animal communication SOCIAL: - Twitter: twitter.com/lexfridman - LinkedIn: linkedin.com/in/lexfridman - Facebook: facebook.com/lexfridman - Instagram: instagram.com/lexfridman - Medium: medium.com/@lexfridman - Reddit: reddit.com/r/lexfridman - Support on Patreon: patreon.com/lexfridman

Why it's here

2 people recommend this

What people say. Recommended for its insights into the human language comprehension network and how it compares to large language models.

photon_lineson Hacker News • Recommends

On “Simple tasks showing reasoning breakdown in state-of-the-art LLMs”

'It's impossible to plan without an internal monologue.' - Sorry, but I disagree with this. I have no 'internal voice' or monologue - whenever I see a problem, my brain actually and fully models it using images. I believe 25% of the population doesn't have the internal monologue which you're referring to and this has been tested and confirmed. I highly recommend listening to this Lex Friedman podcast episode to get a full grasp on the complexities of modelling language and general modelling present in the human brain: link to this video
View on Hacker News

hashemianon Hacker News • Recommends

On “To the brain, reading computer code is not the same as reading language (2020)”

If you are interested in this topic, I suggest watching the conversation between Edward Gibson and Lex Fridman. In the middle of the conversation 1 Edward talks about how there is a "human language comprehension network" within human brain, which gets activated only when we read or speak in any human languages, but nothing else. For example for those speaking multiple languages, reading or writing in any of those languages activates the network. But reading gibberish or computer code, neither activate the network. 1 link to this video
View on Hacker News

Hanschrion Hacker News • Mentions

On “ChatGPT provides false information about people, and OpenAI can't correct it”

…could just as well have been made by a human. I think this lends some credibility to the text produced by the LLM, because more or less all text ever produced has had a human behind it. This is no longer the case and as such there is bound to be a transitional period where we learn how to deal with this new technology. 0: link 1: link 2: link to this video…
View on Hacker News