正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

2. Video-based AI will make text training primitive LeCun's calculation: A 4-year-old processes 10¹⁴ bytes through vision alone. That equals ALL the text used to train GPT-4. In 4 years. Through one sense. This changes everything about how AI should learn:

Karl Mehta

143,679 subscribers

120,950 次观看 • 11 个月前 •via X (Twitter)

健康养生科学技术教育

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

"This is how GPT-4 sees and hears itself" I used GPT-4 to describe itself. Then I used its description to generate an image, a video based on this image and a soundtrack. Tools I used: GPT-4, Midjourney, Kainber AI, Mubert, RunwayML This is the description I used that GPT-4 had of itself as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML.

"This is how GPT-4 sees and hears itself" I used GPT-4 to describe itself. Then I used its description to generate an image, a video based on this image and a soundtrack. Tools I used: GPT-4, Midjourney, Kainber AI, Mubert, RunwayML This is the description I used that GPT-4 had of itself as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML.

Kris Kashtanova

1,233,490 次观看 • 3 年前

Yann LeCun just said something that every AI-in-healthcare researcher should sit with. He basically said: If language were enough to understand the world, you could learn medicine by reading books. But you can’t. You need residency. You need to see thousands of normal cases before you recognize the abnormal one. He also points out something wild — all the public text on the internet is on the order of 10¹⁴ bytes. A 4-year-old processes about that much through vision alone. The world is just… higher bandwidth than text. I think this shift — from language models to world models — is going to matter a lot in healthcare. 🫀

Yann LeCun just said something that every AI-in-healthcare researcher should sit with. He basically said: If language were enough to understand the world, you could learn medicine by reading books. But you can’t. You need residency. You need to see thousands of normal cases before you recognize the abnormal one. He also points out something wild — all the public text on the internet is on the order of 10¹⁴ bytes. A 4-year-old processes about that much through vision alone. The world is just… higher bandwidth than text. I think this shift — from language models to world models — is going to matter a lot in healthcare. 🫀

Bo Wang

418,789 次观看 • 5 个月前

How to humanize AI written text in seconds? 1. Go to 2. Copy and paste your text or upload the file 3. HumanizeThis will humanize your text in seconds 4. After this, it will check it through 8 AI detectors - TurnItIn - GPTZero - Originality AI - CopyLeaks, and so on 5. These detectors confirm the text is humanized 6. You can also see the changes tracked 7. Accept the changes and copy the humanized text The best part? It is FREE Humanizethis Link:

How to humanize AI written text in seconds? 1. Go to 2. Copy and paste your text or upload the file 3. HumanizeThis will humanize your text in seconds 4. After this, it will check it through 8 AI detectors - TurnItIn - GPTZero - Originality AI - CopyLeaks, and so on 5. These detectors confirm the text is humanized 6. You can also see the changes tracked 7. Accept the changes and copy the humanized text The best part? It is FREE Humanizethis Link:

Faheem Ullah

120,426 次观看 • 4 个月前

I searched through over 100+ new AI tools this week. The top 10 most interesting ones that you've never heard about: 1. Genmo Create and edit videos or images from text using GPT-4.

I searched through over 100+ new AI tools this week. The top 10 most interesting ones that you've never heard about: 1. Genmo Create and edit videos or images from text using GPT-4.

Rowan Cheung

1,106,723 次观看 • 3 年前

"This is how GPT-4 sees and hears consciousness" I used GPT-4 to describe consciousness. Then I used its description to generate an image, a video based on this image, and a soundtrack. Tools I used: GPT-4, Midjourney, Kaiber AI, Mubert, RunwayML This is the description I used that GPT-4 had of consciousness as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML: "Consciousness could be visually represented as a complex, multi-dimensional web of interconnected neurons. Imagine a vast, shimmering network of glowing points of light, with each point representing a neuron. These neurons are connected by delicate, pulsating strands of energy, representing the synapses through which information and thoughts flow. This neural network would be constantly evolving and changing, with new connections forming and old ones dissolving, reflecting the dynamic nature of consciousness. The colors of the points and connections could shift through a rich spectrum, symbolizing different emotions, experiences, and cognitive processes. Surrounding this intricate web, there could be a swirling, luminous mist of vibrant colors, representing the ineffable and mysterious nature of consciousness. The mist would ebb and flow, occasionally coalescing into recognizable patterns and forms, hinting at the emergence of thoughts and ideas"

"This is how GPT-4 sees and hears consciousness" I used GPT-4 to describe consciousness. Then I used its description to generate an image, a video based on this image, and a soundtrack. Tools I used: GPT-4, Midjourney, Kaiber AI, Mubert, RunwayML This is the description I used that GPT-4 had of consciousness as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML: "Consciousness could be visually represented as a complex, multi-dimensional web of interconnected neurons. Imagine a vast, shimmering network of glowing points of light, with each point representing a neuron. These neurons are connected by delicate, pulsating strands of energy, representing the synapses through which information and thoughts flow. This neural network would be constantly evolving and changing, with new connections forming and old ones dissolving, reflecting the dynamic nature of consciousness. The colors of the points and connections could shift through a rich spectrum, symbolizing different emotions, experiences, and cognitive processes. Surrounding this intricate web, there could be a swirling, luminous mist of vibrant colors, representing the ineffable and mysterious nature of consciousness. The mist would ebb and flow, occasionally coalescing into recognizable patterns and forms, hinting at the emergence of thoughts and ideas"

Kris Kashtanova

423,380 次观看 • 3 年前

Yann LeCun says all the public text on the internet equals about four years of what a child's eyes take in, and would take you 400,000 years to read: "When you take the totality of all the text available on the internet publicly, it's about 20 trillion words." "That's 30 trillion tokens. A token is two or three bytes. Let's say three. So that's 10 to the 14 bytes." "Of course, the big AI companies are also training with synthetic data and licensed non-public data, but it's not a lot actually compared to all the internet." "10 to the 14 bytes is just about the amount of bytes that gets to your visual cortex in the first four years of your life." "Whereas the text, the publicly available text on the internet would take any of us at least 400,000 years to read through, which is completely impractical obviously." - Yann LeCun (Yann LeCun), co-founder and executive chairman of AMI Labs, at the RAISE Summit (RAISE Summit).

Yann LeCun says all the public text on the internet equals about four years of what a child's eyes take in, and would take you 400,000 years to read: "When you take the totality of all the text available on the internet publicly, it's about 20 trillion words." "That's 30 trillion tokens. A token is two or three bytes. Let's say three. So that's 10 to the 14 bytes." "Of course, the big AI companies are also training with synthetic data and licensed non-public data, but it's not a lot actually compared to all the internet." "10 to the 14 bytes is just about the amount of bytes that gets to your visual cortex in the first four years of your life." "Whereas the text, the publicly available text on the internet would take any of us at least 400,000 years to read through, which is completely impractical obviously." - Yann LeCun (Yann LeCun), co-founder and executive chairman of AMI Labs, at the RAISE Summit (RAISE Summit).

Karl Mehta

23,868 次观看 • 11 天前

.Reiner Pope's new blackboard lecture goes all the way down: how AI training and inference are built up from logic gates on silicon. He walks me through a 4-bit multiply-accumulate by hand, and shows how that primitive is the foundation for the matrix multiplies in training runs.

.Reiner Pope's new blackboard lecture goes all the way down: how AI training and inference are built up from logic gates on silicon. He walks me through a 4-bit multiply-accumulate by hand, and shows how that primitive is the foundation for the matrix multiplies in training runs.

Dwarkesh Patel

38,959 次观看 • 2 个月前

This developer used AI to make David Attenborough narrate his life. A combo of GPT-4-Vision and ElevenLabs' voice cloning generates Attenborough's voice in real-time based on his on-camera movements. It's like being the star of your own documentary.

This developer used AI to make David Attenborough narrate his life. A combo of GPT-4-Vision and ElevenLabs' voice cloning generates Attenborough's voice in real-time based on his on-camera movements. It's like being the star of your own documentary.

Product Hunt 😸

27,514 次观看 • 2 年前

Here's a World War 2 movie trailer I made this weekend, completely generated using AI text-to-video! Script written by GPT-4, Michael Caine voiceover with , and video AI generated with Runway Gen-2 😮 We still have a ways to go but the potential is so exciting!

Here's a World War 2 movie trailer I made this weekend, completely generated using AI text-to-video! Script written by GPT-4, Michael Caine voiceover with , and video AI generated with Runway Gen-2 😮 We still have a ways to go but the potential is so exciting!

Ammaar Reshi

404,941 次观看 • 3 年前

4/ How AI Is Revolutionizing Drug Discovery Scientists had mapped less than 1% of protein structures before DeepMind. Proteins control everything in your body. Mapping ONE used to take years. AI did 200 MILLION in one year. This unlocked thousands of new drugs overnight.

4/ How AI Is Revolutionizing Drug Discovery Scientists had mapped less than 1% of protein structures before DeepMind. Proteins control everything in your body. Mapping ONE used to take years. AI did 200 MILLION in one year. This unlocked thousands of new drugs overnight.

Karl Mehta

74,116 次观看 • 1 年前

Jensen Huang explains that AI has progressed through four major phases. The first wave was Perception AI, where deep learning enabled superhuman vision and speech recognition. The second wave was Generative AI, allowing AI to create text, images, video, and more. We are now in the third wave, Reasoning AI, where models can think through problems, apply logic, and even conduct research. The fourth wave will be Physical AI, where models gain real-world understanding and common sense, leading to advanced robotics.

Jensen Huang explains that AI has progressed through four major phases. The first wave was Perception AI, where deep learning enabled superhuman vision and speech recognition. The second wave was Generative AI, allowing AI to create text, images, video, and more. We are now in the third wave, Reasoning AI, where models can think through problems, apply logic, and even conduct research. The fourth wave will be Physical AI, where models gain real-world understanding and common sense, leading to advanced robotics.

Wes Roth

69,367 次观看 • 1 年前

OpenAI Dev Day Just kicked off and it was the biggest day for AI since the launch of GPT-4 in April. Here's a 3 minute video with everything you need to know and a full summary: 1. GPT-4 Turbo is released. It is much better than GPT-4, can accept up to 128k characters and is 3x cheaper. The knowledge cut off is now April 2023 2. Text to Speech is also being released. They offer 6 hyper-realistic voice models that are the best I've heard from any other AI t-t-s software on the market. 3. Copyright Shield. OpenAI will actually defend your legal costs if you get sued for copyright infringement using their API! 4. GPTs. Maybe the most exciting IMHO, you can now create custom "GPTs" by training your own model of GPT and selling it on their store. This will create a massive shakeup of the current AI software landscape and is the first step towards autonomous agents.

OpenAI Dev Day Just kicked off and it was the biggest day for AI since the launch of GPT-4 in April. Here's a 3 minute video with everything you need to know and a full summary: 1. GPT-4 Turbo is released. It is much better than GPT-4, can accept up to 128k characters and is 3x cheaper. The knowledge cut off is now April 2023 2. Text to Speech is also being released. They offer 6 hyper-realistic voice models that are the best I've heard from any other AI t-t-s software on the market. 3. Copyright Shield. OpenAI will actually defend your legal costs if you get sued for copyright infringement using their API! 4. GPTs. Maybe the most exciting IMHO, you can now create custom "GPTs" by training your own model of GPT and selling it on their store. This will create a massive shakeup of the current AI software landscape and is the first step towards autonomous agents.

Mark Gadala-Maria

79,177 次观看 • 2 年前

ChatGPT for Research No more sifting through endless research papers to find the answer you need. Just ask any question and Consensus AI will deliver a GPT-4-powered summary of the top 5-10 papers.

ChatGPT for Research No more sifting through endless research papers to find the answer you need. Just ask any question and Consensus AI will deliver a GPT-4-powered summary of the top 5-10 papers.

Shubham Saboo

1,142,849 次观看 • 3 年前

How good is GPT-4-Vision at extracting text from images? I wanted to find the limit - but I found weirdness instead Most surprising: GPT-4V performance varies depending on the *structure* of text it sees Let me explain A set of images with progressively more text was presented to GPT-4-Vision. GPT-4V was asked what text it saw in the image. The response from the model was compared against the image’s original text and scored for similarity. The model was tested with 4 types of text: essay, random words, random tokens, and random characters. Findings: * Performance degrades - Yes, the models are good at basic OCR, but as you get more text and words then performance drops (this is expected) * Type of context matters - You should expect different recall on your texts based on your context types * Hallucination Errors - I thought that the model would make errors of omission (it wouldn’t return all the words). But instead the model mostly made hallucination errors - it replaced words with made up words. * Evals Matter - This test in isolation doesn’t mean that your data will have the same results, but it should motivate you to create eval tests for your data and anticipate errors which are hard to spot Notes: * Next step would be to add additional image types like tables or PDFs * GPT-4V would routinely get stuck in repeat-token-loops when trying to extract random tokens * GPT-4V would refuse to answer most random character images

How good is GPT-4-Vision at extracting text from images? I wanted to find the limit - but I found weirdness instead Most surprising: GPT-4V performance varies depending on the structure of text it sees Let me explain A set of images with progressively more text was presented to GPT-4-Vision. GPT-4V was asked what text it saw in the image. The response from the model was compared against the image’s original text and scored for similarity. The model was tested with 4 types of text: essay, random words, random tokens, and random characters. Findings: * Performance degrades - Yes, the models are good at basic OCR, but as you get more text and words then performance drops (this is expected) * Type of context matters - You should expect different recall on your texts based on your context types * Hallucination Errors - I thought that the model would make errors of omission (it wouldn’t return all the words). But instead the model mostly made hallucination errors - it replaced words with made up words. * Evals Matter - This test in isolation doesn’t mean that your data will have the same results, but it should motivate you to create eval tests for your data and anticipate errors which are hard to spot Notes: * Next step would be to add additional image types like tables or PDFs * GPT-4V would routinely get stuck in repeat-token-loops when trying to extract random tokens * GPT-4V would refuse to answer most random character images

Greg Kamradt

49,111 次观看 • 2 年前

How to Analyze Tables In Large Financial Reports Using GPT-4 (w/Jerry Liu) Most corporate docs contain a mix of text and tables. But if you use simple RAG split and chunk methods, the AI model will likely hallucinate due to embedding split tables. To solve this, Jerry Liu demonstrates how to use LlamaIndex 🦙,OpenAI's new GPT-1106 model, and a novel retrieval strategy to analyse financial statement tables in Tesla's 10-k. Full tutorial video:

How to Analyze Tables In Large Financial Reports Using GPT-4 (w/Jerry Liu) Most corporate docs contain a mix of text and tables. But if you use simple RAG split and chunk methods, the AI model will likely hallucinate due to embedding split tables. To solve this, Jerry Liu demonstrates how to use LlamaIndex 🦙,OpenAI's new GPT-1106 model, and a novel retrieval strategy to analyse financial statement tables in Tesla's 10-k. Full tutorial video:

Mayo Oshin

185,199 次观看 • 2 年前

GPT-4 + AR = empowering users to interact with real-world objects and devices through AI. Introducing Fusion Reality: You can ask AI to control a light, check the weather, or inquire about the medicine you're holding and its instructions in AR.

GPT-4 + AR = empowering users to interact with real-world objects and devices through AI. Introducing Fusion Reality: You can ask AI to control a light, check the weather, or inquire about the medicine you're holding and its instructions in AR.

Xiaoan Liu

59,010 次观看 • 3 年前

"This is how GPT-4 sees and hears the human soul" I used GPT-4 to describe the human soul. Then I used its description to generate an image, a video based on this image, and a soundtrack. Tools I used: GPT-4, Midjourney, Kaiber AI, Mubert, RunwayML This is the description I used that GPT-4 had of the human soul as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML: "Imagine a translucent, radiant figure in the shape of a human body, composed of shimmering, colorful energy representing an individual's unique character and personality. This figure is surrounded by a soft, glowing aura that reflects the person's spiritual essence. Inside the figure's head, a small, luminous orb is suspended, symbolizing the seat of consciousness, self-awareness, and decision-making. This orb is connected to the rest of the radiant body by threads of light, representing the interplay between the physical and spiritual realms. The figure appears to be gently floating above the ground, evoking the eternal aspect of the soul and its ability to transcend the physical world. The background could feature various symbols or representations of the afterlife, reincarnation, or merging with a higher power, highlighting the diverse spiritual and religious beliefs about the soul's journey beyond the physical body."

"This is how GPT-4 sees and hears the human soul" I used GPT-4 to describe the human soul. Then I used its description to generate an image, a video based on this image, and a soundtrack. Tools I used: GPT-4, Midjourney, Kaiber AI, Mubert, RunwayML This is the description I used that GPT-4 had of the human soul as a prompt to text-to-image, image-to-video, and text-to-music. I put the video and sound together in RunwayML: "Imagine a translucent, radiant figure in the shape of a human body, composed of shimmering, colorful energy representing an individual's unique character and personality. This figure is surrounded by a soft, glowing aura that reflects the person's spiritual essence. Inside the figure's head, a small, luminous orb is suspended, symbolizing the seat of consciousness, self-awareness, and decision-making. This orb is connected to the rest of the radiant body by threads of light, representing the interplay between the physical and spiritual realms. The figure appears to be gently floating above the ground, evoking the eternal aspect of the soul and its ability to transcend the physical world. The background could feature various symbols or representations of the afterlife, reincarnation, or merging with a higher power, highlighting the diverse spiritual and religious beliefs about the soul's journey beyond the physical body."

Kris Kashtanova

465,334 次观看 • 3 年前

I built a service desk agent using ElevenLabs’ new Conversational AI Agents feature. Watch the video to see how responsive it is! Previously, I used ElevenLabs for cloning my voice and used my generated voice for narrations in some of my youtube videos. This feature takes elevenlabs' voice AI to a whole new level! It simplifies systems that used to require separate TTS (text-to-speech) and STT (speech-to-text) processes for both sides of the conversation. Now, it’s much simpler! Create your own agent here What will you create with this?

I built a service desk agent using ElevenLabs’ new Conversational AI Agents feature. Watch the video to see how responsive it is! Previously, I used ElevenLabs for cloning my voice and used my generated voice for narrations in some of my youtube videos. This feature takes elevenlabs' voice AI to a whole new level! It simplifies systems that used to require separate TTS (text-to-speech) and STT (speech-to-text) processes for both sides of the conversation. Now, it’s much simpler! Create your own agent here What will you create with this?

Melvin Vivas

27,722 次观看 • 1 年前

Yann LeCun explains that large language models are trained on about 30 trillion words, representing nearly all public internet text. He says it would take a human over 500,000 years to read that much. But a 4-year-old child sees just as much visual data in their first few years of life. This shows how much richer and more complex real-world experience is compared to reading text. Training on the web is huge but it still doesn’t match what a child learns just by living.

Yann LeCun explains that large language models are trained on about 30 trillion words, representing nearly all public internet text. He says it would take a human over 500,000 years to read that much. But a 4-year-old child sees just as much visual data in their first few years of life. This shows how much richer and more complex real-world experience is compared to reading text. Training on the web is huge but it still doesn’t match what a child learns just by living.

Wes Roth

294,390 次观看 • 7 个月前

More experiments of this workflow: Generating quick 3d models using Lumas Genie 3D. A text prompt generates 4 models in 10 seconds. Then, using Lumas webviewer, I feed a screencapture of the model into Krea ai which details the model based on another prompt. Finally, Magnific AI is used to add the final finish. The Santa mechs are insanely refined😱 Check out the final "Renders" below. #3d #ai

More experiments of this workflow: Generating quick 3d models using Lumas Genie 3D. A text prompt generates 4 models in 10 seconds. Then, using Lumas webviewer, I feed a screencapture of the model into Krea ai which details the model based on another prompt. Finally, Magnific AI is used to add the final finish. The Santa mechs are insanely refined😱 Check out the final "Renders" below. #3d #ai

Martin Nebelong

88,080 次观看 • 2 年前