Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

New work on LLM Control Theory! We frame LLMs as discrete dynamical systems and study prompting through the lens of reachability and controllability. We even proved some theorems!! Joint work my fellow EngSci 2T2 Cameron Witkowski, Dr. @Sz_Looi, advised by Prof. Matt Thomson 🚀

ABhargava2000's profile picture

Aman Bhargava

1,157 subscribers

71,121 Aufrufe • vor 2 Jahren •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

We just dropped our main show on "What's the Magic Word? A Control Theory of LLM Prompting". This is some of the bonus footage we didn't release on YT. With Aman Bhargava Cameron Witkowski @Sz_Looi
11:06

We just dropped our main show on "What's the Magic Word? A Control Theory of LLM Prompting". This is some of the bonus footage we didn't release on YT. With Aman Bhargava Cameron Witkowski @Sz_Looi

Machine Learning Street Talk

21,697 Aufrufe • vor 2 Jahren

New work with Yide Shentu! We present Latent Code as Bridges🌉 or LCB. We leverage LLMs for robotics by finetuning the LLM to output latent embeddings to control the lower level policy. This hierarchical approach enables end2end training while maintaining the advantages of LLMs!
0:24

New work with Yide Shentu! We present Latent Code as Bridges🌉 or LCB. We leverage LLMs for robotics by finetuning the LLM to output latent embeddings to control the lower level policy. This hierarchical approach enables end2end training while maintaining the advantages of LLMs!

Philipp Wu

21,072 Aufrufe • vor 2 Jahren

New work on Geometric Clifford Algebra Networks (GCANs). We propose geometric templates for modeling dynamical systems. A 🧵on geometric / Clifford algebras, and symmetry group transformations in neural networks. 📜
0:23

New work on Geometric Clifford Algebra Networks (GCANs). We propose geometric templates for modeling dynamical systems. A 🧵on geometric / Clifford algebras, and symmetry group transformations in neural networks. 📜

David Ruhe

137,973 Aufrufe • vor 3 Jahren

LLMs have complex joint beliefs about all sorts of quantities. And my postdoc James Requeima visualized them! In this thread we show LLM predictive distributions conditioned on data and free-form text. LLMs pick up on all kinds of subtle and unusual structure: 🧵
0:26

LLMs have complex joint beliefs about all sorts of quantities. And my postdoc James Requeima visualized them! In this thread we show LLM predictive distributions conditioned on data and free-form text. LLMs pick up on all kinds of subtle and unusual structure: 🧵

David Duvenaud

193,619 Aufrufe • vor 1 Jahr

Celebrating the artistry of illustration, we reflect on the work of talented individuals who reimagined Tiffany & Co. icons through a lens of craft and creativity. #TiffanyAndCo #ATiffanyHoliday
1:20

Celebrating the artistry of illustration, we reflect on the work of talented individuals who reimagined Tiffany & Co. icons through a lens of craft and creativity. #TiffanyAndCo #ATiffanyHoliday

Tiffany & Co.

12,428 Aufrufe • vor 6 Monaten

Should the right support Matt Goodwin in the Gorton and Denton by-election? Some have become suspicious of Matt's conversion: from fighting Islamophobia for the Cameron government to standing as a Reform candidate to stop white Brits becoming a minority in their own country. What does Matt Goodwin believe? I went through Matt's written work and recent research on last night's stream. I believe Matt is sincere; and if we want Reform to be better ahead of forming the next government, then it benefits us to have Matt Goodwin as an MP — possibly, even as Reform's Shadow Home Secretary.
2:19:16

Should the right support Matt Goodwin in the Gorton and Denton by-election? Some have become suspicious of Matt's conversion: from fighting Islamophobia for the Cameron government to standing as a Reform candidate to stop white Brits becoming a minority in their own country. What does Matt Goodwin believe? I went through Matt's written work and recent research on last night's stream. I believe Matt is sincere; and if we want Reform to be better ahead of forming the next government, then it benefits us to have Matt Goodwin as an MP — possibly, even as Reform's Shadow Home Secretary.

Connor Tomlinson

47,428 Aufrufe • vor 4 Monaten

New course! Generative AI with Large Language Models, created with Amazon Web Services and hosted on Coursera. This course goes deep into the technical foundations of LLMs and how to use them. You can sign up here: You’ll work through the full life-cycle of a generative AI project, and learn specific techniques like RLHF; zero-shot, one-shot, and few-shot learning with LLMs; advanced prompting frameworks like ReAct; even fine-tuning LLMs, and gain hands-on practice with all of these techniques. Instructors Antje Barth Chris Fregly Shelbee Eigenbrode and Mike G Chambers all do incredible Generative AI work at AWS, and have supported many companies to build creative LLM applications. They bring tremendous practical LLM expertise to this course. I'm confident you’ll finish this course with a deeper understanding of how LLMs work, and how to use them. I hope you enjoy the course!
1:57

New course! Generative AI with Large Language Models, created with Amazon Web Services and hosted on Coursera. This course goes deep into the technical foundations of LLMs and how to use them. You can sign up here: You’ll work through the full life-cycle of a generative AI project, and learn specific techniques like RLHF; zero-shot, one-shot, and few-shot learning with LLMs; advanced prompting frameworks like ReAct; even fine-tuning LLMs, and gain hands-on practice with all of these techniques. Instructors Antje Barth Chris Fregly Shelbee Eigenbrode and Mike G Chambers all do incredible Generative AI work at AWS, and have supported many companies to build creative LLM applications. They bring tremendous practical LLM expertise to this course. I'm confident you’ll finish this course with a deeper understanding of how LLMs work, and how to use them. I hope you enjoy the course!

Andrew Ng

467,875 Aufrufe • vor 3 Jahren

Senator Tuberville: We Don't Work For The Damn Senate, We Work For The People Of This Country, And If We Don't Do Something Now It Will Never Get Done. We Will Lose And Our Kids, And Grandkids, And Even Us Will Be Punished By The Democrats When They Take Control Again. Coach Tommy Tuberville
4:23

Senator Tuberville: We Don't Work For The Damn Senate, We Work For The People Of This Country, And If We Don't Do Something Now It Will Never Get Done. We Will Lose And Our Kids, And Grandkids, And Even Us Will Be Punished By The Democrats When They Take Control Again. Coach Tommy Tuberville

Bannon’s WarRoom

41,187 Aufrufe • vor 5 Monaten

🔥 NEW $WORK LISTING DEC 18TH 🔥 📅 18th of Dec - 3PM CET 🌏 Work X will continue to expand the availability of the $WORK token globally and across several eco-systems🚀🚀
0:20

🔥 NEW $WORK LISTING DEC 18TH 🔥 📅 18th of Dec - 3PM CET 🌏 Work X will continue to expand the availability of the $WORK token globally and across several eco-systems🚀🚀

Echo Agent

189,076 Aufrufe • vor 2 Jahren

When we work on finalizing rules U.S. Securities and Exchange Commission, we make sure to have an implementation phase and we help folks manage through the implementation by answering questions and trying to facilitate the work of the private sector to implement properly.
1:35

When we work on finalizing rules U.S. Securities and Exchange Commission, we make sure to have an implementation phase and we help folks manage through the implementation by answering questions and trying to facilitate the work of the private sector to implement properly.

SEC Chair Gary Gensler Archive

58,573 Aufrufe • vor 2 Jahren

Get me on your sets. This is some of my finest work. Also new And Now We Drink with Derek Kage
2:37

Get me on your sets. This is some of my finest work. Also new And Now We Drink with Derek Kage

Matt Slayer

27,999 Aufrufe • vor 1 Jahr

Because of my father’s story, I define my whole outlook of America through the lens of the American Dream and immigration. That’s why it’s so painful how politicized it’s become, but I know if we work together, we can solve this problem and many more.
1:05

Because of my father’s story, I define my whole outlook of America through the lens of the American Dream and immigration. That’s why it’s so painful how politicized it’s become, but I know if we work together, we can solve this problem and many more.

Tom Suozzi

46,310 Aufrufe • vor 1 Jahr

KAMALA: "We will work together, and continue to work together, to address these issues, to tackle these challenges, and to work together as we continue to work operating from the new norms, rules, and agreements, that we will convene to work together. We will work together." 🤦‍♂️
0:34

KAMALA: "We will work together, and continue to work together, to address these issues, to tackle these challenges, and to work together as we continue to work operating from the new norms, rules, and agreements, that we will convene to work together. We will work together." 🤦‍♂️

Cillian

649,127 Aufrufe • vor 1 Jahr

🐟 some musings on how we might use LLMs 🐠 to interact with text at multiple levels of abstraction 🐡 inspired by the fish-eye lens
1:24

🐟 some musings on how we might use LLMs 🐠 to interact with text at multiple levels of abstraction 🐡 inspired by the fish-eye lens

Amelia Wattenberger 🪷

61,905 Aufrufe • vor 1 Jahr

*Finally* read through Sam Rose's blog on LLM quantization. It's incredible. For many (even in tech) the understanding of how LLMs work stops at the surface level. Sam is helping us all go deeper, digging into the interesting facets of how AI models truly work. Read it!
0:23

*Finally* read through Sam Rose's blog on LLM quantization. It's incredible. For many (even in tech) the understanding of how LLMs work stops at the surface level. Sam is helping us all go deeper, digging into the interesting facets of how AI models truly work. Read it!

Ben Dicken

272,292 Aufrufe • vor 2 Monaten

“We consider the research work and findings put out by Global InfoAnalytics as an opportunity and feedback to improve the work we are doing”
1:33

“We consider the research work and findings put out by Global InfoAnalytics as an opportunity and feedback to improve the work we are doing”

Kwasi Kwarteng

40,852 Aufrufe • vor 10 Monaten

Reasoning LLMs Guide [Full Video - Unedited] 1 hr talk on reasoning LLMs and how to best use them for different applications. Share with your devs & students. I discuss lots of fun ideas like meta-prompting, LLM-as-a-Judge, use cases, prompting tips, and much more.
1:05:07

Reasoning LLMs Guide [Full Video - Unedited] 1 hr talk on reasoning LLMs and how to best use them for different applications. Share with your devs & students. I discuss lots of fun ideas like meta-prompting, LLM-as-a-Judge, use cases, prompting tips, and much more.

elvis

61,139 Aufrufe • vor 1 Jahr

Legacy media doesn't like being on the other side of the lens. They're arms of the establishment and will work in cahoots to destroy independent journalists, as we are the media now.
5:07

Legacy media doesn't like being on the other side of the lens. They're arms of the establishment and will work in cahoots to destroy independent journalists, as we are the media now.

Tommy Robinson 🇬🇧

247,397 Aufrufe • vor 1 Jahr

🚀 Introducing Cover-Agent 🧪 An open-source tool that includes a reimplementation of Meta's TestGen-LLM for automatically enhancing test suites. Manager: "We must improve old test suites for better code coverage. Can you handle it?" Me: "Sure, my favorite task... (Not!) 🤷‍♂️" Meta's team had the idea of using LLMs to enhance test suites—sounds great, right? But they didn't release their code We did. We'd love your feedback as we work on improving this baseline In this video, I'll review TestGen-LLM, share insights, and introduce you to Cover-Agent
6:11

🚀 Introducing Cover-Agent 🧪 An open-source tool that includes a reimplementation of Meta's TestGen-LLM for automatically enhancing test suites. Manager: "We must improve old test suites for better code coverage. Can you handle it?" Me: "Sure, my favorite task... (Not!) 🤷‍♂️" Meta's team had the idea of using LLMs to enhance test suites—sounds great, right? But they didn't release their code We did. We'd love your feedback as we work on improving this baseline In this video, I'll review TestGen-LLM, share insights, and introduce you to Cover-Agent

Itamar Friedman

139,228 Aufrufe • vor 2 Jahren