Video yükleniyor...

Video Yüklenemedi

Bu video yüklenirken bir sorun oluştu. Bu geçici bir ağ sorunundan kaynaklanıyor olabilir veya video kullanılamıyor olabilir.

Ana Sayfaya Dön

Francois Chollet says LLMs aren't enough for human-like continual learning They store skills as vector programs learned via gradient descent — not efficient, not adaptive True AGI needs to learn from experience and generalize fast "LLMs can be part of AGI, but not the solution"

Haider.

66,939 subscribers

50,251 görüntüleme • 8 ay önce •via X (Twitter)

Eğitim Bilim & Teknoloji

Anya Rossi• Live Now

Private livecam show

0 Yorum

Yorum bulunmuyor

Orijinal gönderinin yorumları burada görünecek

Benzer Videolar

Demis Hassabis says large language models (LLMs) can search and plan but they forget everything once the session ends. This is because today’s models don’t truly “learn” after deployment. He sees continual learning and better memory as key missing ingredients for AGI. Efficient memory, like the brain’s ability to store only what matters, will be crucial. LLMs are powerful, but without memory, they have a "goldfish brain."

Demis Hassabis says large language models (LLMs) can search and plan but they forget everything once the session ends. This is because today’s models don’t truly “learn” after deployment. He sees continual learning and better memory as key missing ingredients for AGI. Efficient memory, like the brain’s ability to store only what matters, will be crucial. LLMs are powerful, but without memory, they have a "goldfish brain."

Wes Roth

38,833 görüntüleme • 5 ay önce

Francois Chollet: OpenAI has set back the progress towards AGI by 5-10 years because frontier research is no longer being published and LLMs are an offramp on the path to AGI

Francois Chollet: OpenAI has set back the progress towards AGI by 5-10 years because frontier research is no longer being published and LLMs are an offramp on the path to AGI

Tsarathustra

889,085 görüntüleme • 2 yıl önce

David Deutsch says all computation from humans to LLMs shares the same fundamental power: the universal Turing machine The difference is only in programs and outputs. But AGI is different; we don't even understand it in principle. "so the true sign would be a theory, not a machine"

David Deutsch says all computation from humans to LLMs shares the same fundamental power: the universal Turing machine The difference is only in programs and outputs. But AGI is different; we don't even understand it in principle. "so the true sign would be a theory, not a machine"

Haider.

11,720 görüntüleme • 10 ay önce

Amjad Masad says we may have “functional AGI,” not true AGI: automation that works because humans supply the expertise. True AGI would learn new skills on the fly. For now, humans are still the teachers. Amjad Masad

Amjad Masad says we may have “functional AGI,” not true AGI: automation that works because humans supply the expertise. True AGI would learn new skills on the fly. For now, humans are still the teachers. Amjad Masad

a16z

144,888 görüntüleme • 8 ay önce

this was the most important bit of the ilya interview imo. humans are not AGI, and LLMs won’t become like humans with more scaling. humans lack a huge amount of knowledge and have insanely good “continual learning” powered by extremely good reward functions (except for addiction) that guide them. and so it’s more about the learning rather than knowledge. the term AGI was coined as a response to the narrow AI which was only capable of intelligence in a specific domain (chess, game, code, etc) but General intelligence is able to learn different topics because that’s what “learning” is about. pre-training was thought to be the solution to reaching AGI because you train AI on so many different topics. but they have “overshot” that target. amazing perspective.

this was the most important bit of the ilya interview imo. humans are not AGI, and LLMs won’t become like humans with more scaling. humans lack a huge amount of knowledge and have insanely good “continual learning” powered by extremely good reward functions (except for addiction) that guide them. and so it’s more about the learning rather than knowledge. the term AGI was coined as a response to the narrow AI which was only capable of intelligence in a specific domain (chess, game, code, etc) but General intelligence is able to learn different topics because that’s what “learning” is about. pre-training was thought to be the solution to reaching AGI because you train AI on so many different topics. but they have “overshot” that target. amazing perspective.

ℏεsam

53,427 görüntüleme • 7 ay önce

My new podcast. Links below. First episode: Are LLMs creative? 0:25 LLM outputs are implicit in the weights 2:10 All programs are inherently limited 3:51 There is an AGI program 7:12 LLMs are not ‘just math’ 8:05 LLMs are more than just the training data 9:07 Can LLMs be as creative as DNA? 12:12 Q&A

My new podcast. Links below. First episode: Are LLMs creative? 0:25 LLM outputs are implicit in the weights 2:10 All programs are inherently limited 3:51 There is an AGI program 7:12 LLMs are not ‘just math’ 8:05 LLMs are more than just the training data 9:07 Can LLMs be as creative as DNA? 12:12 Q&A

Nivi

20,382 görüntüleme • 2 yıl önce

Continual Learning in 2026. And AGI with it? In July, Dario said he believes Continual Learning will turn out to be not as difficult as it seems. Sholto Douglas, Anthropic and former Deepmind researcher, predicted Continual Learning gets solved in a satisfying way in 2026. Sholto said that last week. Considering how capable AI models have become in 2025, it is unwise to dismiss this as fundraising or hype. It is much wiser to assume AGI is coming. That it will change the world. And that you can benefit from it... ...than it is to assume the opposite. "Not enough people are ready for AI to not be a bubble." - roon Choose wisely my friends.

Continual Learning in 2026. And AGI with it? In July, Dario said he believes Continual Learning will turn out to be not as difficult as it seems. Sholto Douglas, Anthropic and former Deepmind researcher, predicted Continual Learning gets solved in a satisfying way in 2026. Sholto said that last week. Considering how capable AI models have become in 2025, it is unwise to dismiss this as fundraising or hype. It is much wiser to assume AGI is coming. That it will change the world. And that you can benefit from it... ...than it is to assume the opposite. "Not enough people are ready for AI to not be a bubble." - roon Choose wisely my friends.

Dan McAteer

69,496 görüntüleme • 6 ay önce

Geoffrey Hinton says LLMs work like the human brain, not like normal computer software They aren't built on explainable lines of code, but rather on trillions of connection strengths that learn from data "it's largely a mystery" We don't know what the individual neurons are doing

Geoffrey Hinton says LLMs work like the human brain, not like normal computer software They aren't built on explainable lines of code, but rather on trillions of connection strengths that learn from data "it's largely a mystery" We don't know what the individual neurons are doing

Haider.

35,283 görüntüleme • 5 ay önce

Francois Chollet + Sam Altman Fireside François Chollet and Sam Altman fireside during ARC-AGI-3 Launch Party moderated by Deedy They discuss: - Social contracts evolving - AGI views as a parent - When will labs score >85% on ARC-AGI-3?

Francois Chollet + Sam Altman Fireside François Chollet and Sam Altman fireside during ARC-AGI-3 Launch Party moderated by Deedy They discuss: - Social contracts evolving - AGI views as a parent - When will labs score >85% on ARC-AGI-3?

ARC Prize

33,317 görüntüleme • 3 ay önce

François Chollet says most definitions of AGI are really definitions of automation. AGI isn’t about doing tasks. It’s about learning new ones as efficiently as a human. We may automate the economy before we build intelligence.

François Chollet says most definitions of AGI are really definitions of automation. AGI isn’t about doing tasks. It’s about learning new ones as efficiently as a human. We may automate the economy before we build intelligence.

vitrupo

11,327 görüntüleme • 2 ay önce

🚨 FRANCOIS CHOLLET SAYS THE CURRENT LLM STACK IS THE WRONG PATH TO AGI Interviewer: “When will we accomplish the first definition of AGI?” Chollet confirms we are already on that exact trajectory, pointing out that: "Current technology can fully automate at human level or beyond any domain where you have verifiable rewards, right? And code... code being the first one." He then brings up the problem with LLMs, stating that: "It's possible in principle to build something that looks a lot like AGI on top of the LLM stack... I do believe, however, this would be the wrong thing to do because it would be very inefficient." Looking ahead to what the final architecture will actually require, he explains that: "AI research will have to trend towards not just efficiency, but in fact optimality over time. And for this reason, future AI in a few decades... it's not going to be this harness on top of a reasoning model on top of a base LLM."

🚨 FRANCOIS CHOLLET SAYS THE CURRENT LLM STACK IS THE WRONG PATH TO AGI Interviewer: “When will we accomplish the first definition of AGI?” Chollet confirms we are already on that exact trajectory, pointing out that: "Current technology can fully automate at human level or beyond any domain where you have verifiable rewards, right? And code... code being the first one." He then brings up the problem with LLMs, stating that: "It's possible in principle to build something that looks a lot like AGI on top of the LLM stack... I do believe, however, this would be the wrong thing to do because it would be very inefficient." Looking ahead to what the final architecture will actually require, he explains that: "AI research will have to trend towards not just efficiency, but in fact optimality over time. And for this reason, future AI in a few decades... it's not going to be this harness on top of a reasoning model on top of a base LLM."

Chris

15,578 görüntüleme • 2 ay önce

Computer scientist Judea Pearl: There are mathematical limits to LLMs that cannot be crossed by scaling alone LLMs don't discover world models from raw data; they merely summarize the interpretations humans have already written down "this path is not the way to get AGI"

Computer scientist Judea Pearl: There are mathematical limits to LLMs that cannot be crossed by scaling alone LLMs don't discover world models from raw data; they merely summarize the interpretations humans have already written down "this path is not the way to get AGI"

Haider.

622,755 görüntüleme • 5 ay önce

François Chollet: AGI is likely by 2030–early 2030s, based on current progress and investment across LLMs, plus side bets that might work out By the time ARC-AGI reaches version 6 or 7, we'll probably have AGI

François Chollet: AGI is likely by 2030–early 2030s, based on current progress and investment across LLMs, plus side bets that might work out By the time ARC-AGI reaches version 6 or 7, we'll probably have AGI

Haider.

15,947 görüntüleme • 3 ay önce

.Richard Sutton, father of reinforcement learning, doesn’t think LLMs are bitter-lesson-pilled. My steel man of Richard’s position: we need some new architecture to enable continual (on-the-job) learning. And if we have continual learning, we don't need a special training phase - the agent just learns on-the-fly - like all humans, and indeed, like all animals. This new paradigm will render our current approach with LLMs obsolete. I did my best to represent the view that LLMs will function as the foundation on which this experiential learning can happen. Some sparks flew. 0:00:00 – Are LLMs a dead-end? 0:13:51 – Do humans do imitation learning? 0:23:57 – The Era of Experience 0:34:25 – Current architectures generalize poorly out of distribution 0:42:17 – Surprises in the AI field 0:47:28 – Will The Bitter Lesson still apply after AGI? 0:54:35 – Succession to AI

.Richard Sutton, father of reinforcement learning, doesn’t think LLMs are bitter-lesson-pilled. My steel man of Richard’s position: we need some new architecture to enable continual (on-the-job) learning. And if we have continual learning, we don't need a special training phase - the agent just learns on-the-fly - like all humans, and indeed, like all animals. This new paradigm will render our current approach with LLMs obsolete. I did my best to represent the view that LLMs will function as the foundation on which this experiential learning can happen. Some sparks flew. 0:00:00 – Are LLMs a dead-end? 0:13:51 – Do humans do imitation learning? 0:23:57 – The Era of Experience 0:34:25 – Current architectures generalize poorly out of distribution 0:42:17 – Surprises in the AI field 0:47:28 – Will The Bitter Lesson still apply after AGI? 0:54:35 – Succession to AI

Dwarkesh Patel

3,080,088 görüntüleme • 9 ay önce

Google DeepMind's Logan Kilpatrick says AGI will be a product experience. Not a model. His bet: whoever nails memory + context around decent model at a product level wins. Users will suddenly feel like they're talking to AGI. Not from capability breakthrough, but experience breakthrough. At some point, the AGI race is actually a UX problem.

Google DeepMind's Logan Kilpatrick says AGI will be a product experience. Not a model. His bet: whoever nails memory + context around decent model at a product level wins. Users will suddenly feel like they're talking to AGI. Not from capability breakthrough, but experience breakthrough. At some point, the AGI race is actually a UX problem.

vitrupo

177,073 görüntüleme • 1 yıl önce

We just released François Chollet keynote talk from #AGI24. He challenges LLM hype, discusses ARC-AGI benchmark, and proposes merging deep learning with program synthesis for true AGI progress.

We just released François Chollet keynote talk from #AGI24. He challenges LLM hype, discusses ARC-AGI benchmark, and proposes merging deep learning with program synthesis for true AGI progress.

Machine Learning Street Talk

178,245 görüntüleme • 1 yıl önce

We can teach LLMs to write better robot code through natural language feedback. But can LLMs remember what they were taught and improve their teachability over time? Introducing our latest work, Learning to Learn Faster from Human Feedback with Language Model Predictive Control

We can teach LLMs to write better robot code through natural language feedback. But can LLMs remember what they were taught and improve their teachability over time? Introducing our latest work, Learning to Learn Faster from Human Feedback with Language Model Predictive Control

Jacky Liang

86,652 görüntüleme • 2 yıl önce

Journalist Karen Hao says AGI is not scientifically grounded; it's a quasi-religious movement in Silicon Valley Since we still don't agree on what human intelligence is, companies like OpenAI can shape the definition of AI to suit their needs "the belief that AGI is coming soon is faith, not evidence"

Journalist Karen Hao says AGI is not scientifically grounded; it's a quasi-religious movement in Silicon Valley Since we still don't agree on what human intelligence is, companies like OpenAI can shape the definition of AI to suit their needs "the belief that AGI is coming soon is faith, not evidence"

Haider.

411,790 görüntüleme • 6 ay önce

Google DeepMind say true AGI will “reason, adapt, and learn continuously,” and they estimate it’s about 5-10 years away. Meta’s chief scientist Yann LeCun likewise views continual learning as a key pillar of human level AI and is optimistic about achieving it by around 2030 Currently, large language models only improve via offline updates (fine-tuning or reinforcement learning from human feedback) rather than meta learning or modular networks that update themselves gradually. Thankfully All major research teams (OpenAI, DeepMind, Anthropic, Meta, etc.) are actively working on these directions, aiming for models that continuously learn from mistakes as they interact a capability they believe will emerge as we approach true AGI later this decade 2029

Google DeepMind say true AGI will “reason, adapt, and learn continuously,” and they estimate it’s about 5-10 years away. Meta’s chief scientist Yann LeCun likewise views continual learning as a key pillar of human level AI and is optimistic about achieving it by around 2030 Currently, large language models only improve via offline updates (fine-tuning or reinforcement learning from human feedback) rather than meta learning or modular networks that update themselves gradually. Thankfully All major research teams (OpenAI, DeepMind, Anthropic, Meta, etc.) are actively working on these directions, aiming for models that continuously learn from mistakes as they interact a capability they believe will emerge as we approach true AGI later this decade 2029

Chris

284,188 görüntüleme • 9 ay önce