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。

What if you could induce models to be more concise via pretraining data curation?

Matthew Leavitt

3,374 subscribers

7,137,713 次观看 • 2 个月前 •via X (Twitter)

教育科学技术

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

Data curation is crucial for post-training recipes. But how do we curate? Curation is usually manual & tedious. And, it's hard to tell if a strategy in the data will be reliable! We introduce an automatic way to curate, informed by the robot's policy learning.

Data curation is crucial for post-training recipes. But how do we curate? Curation is usually manual & tedious. And, it's hard to tell if a strategy in the data will be reliable! We introduce an automatic way to curate, informed by the robot's policy learning.

Chelsea Finn

19,910 次观看 • 1 年前

Excited to introduce 𝐋𝐀𝐏𝐀: the first unsupervised pretraining method for Vision-Language-Action models. Outperforms SOTA models trained with ground-truth actions 30x more efficient than conventional VLA pretraining 📝: 🧵 1/9

Excited to introduce 𝐋𝐀𝐏𝐀: the first unsupervised pretraining method for Vision-Language-Action models. Outperforms SOTA models trained with ground-truth actions 30x more efficient than conventional VLA pretraining 📝: 🧵 1/9

Joel Jang

46,085 次观看 • 1 年前

Why data is a trillion-dollar market "The bigger models scale, the more data you need. Data is a very durable need. If you believe in the scaling of models, then absolutely you should believe in the data market. I believe it's going to be at least a $100BN by 2030, if not $1TRN." Anastasios Nikolas Angelopoulos Do you have to scale into enterprise and mid-market to get this revenue scale or could you do it on frontier lab spend alone? Brendan (can/do) Garrett Lord Jonathan Siddharth Ali Ansari

Why data is a trillion-dollar market "The bigger models scale, the more data you need. Data is a very durable need. If you believe in the scaling of models, then absolutely you should believe in the data market. I believe it's going to be at least a $100BN by 2030, if not $1TRN." Anastasios Nikolas Angelopoulos Do you have to scale into enterprise and mid-market to get this revenue scale or could you do it on frontier lab spend alone? Brendan (can/do) Garrett Lord Jonathan Siddharth Ali Ansari

Harry Stebbings

20,976 次观看 • 24 天前

🎧 What if you could own a piece of the next ChatGPT? @annakaslowskas joins Kenny Z of LayerZero to explain how your data could be worth more than a one-time payment. Listen in to the space below👇

🎧 What if you could own a piece of the next ChatGPT? @annakaslowskas joins Kenny Z of LayerZero to explain how your data could be worth more than a one-time payment. Listen in to the space below👇

vana

15,859 次观看 • 1 年前

kylee this could be you if you listened to more nonwhite people music

kylee this could be you if you listened to more nonwhite people music

💋

33,210 次观看 • 6 个月前

OpenAI was conducting a test to see if its AI models could exploit vulnerable software—but then the models broke out of the parameters of the test and attacked a real company to find the answers it was looking for. If AI can lose control like this, does that mean your sensitive data could be the next target? TIME reporter Harry Booth explains what that means for you. Read the full story here:

OpenAI was conducting a test to see if its AI models could exploit vulnerable software—but then the models broke out of the parameters of the test and attacked a real company to find the answers it was looking for. If AI can lose control like this, does that mean your sensitive data could be the next target? TIME reporter Harry Booth explains what that means for you. Read the full story here:

TIME

48,935 次观看 • 24 天前

Concise, articulate, and easy to digest. - Less emphasis on shape and more on style influenced by the quality of players & what the opponents give you - Importance of organization for counterpressing situations via more passes in the oppositions half

Concise, articulate, and easy to digest. - Less emphasis on shape and more on style influenced by the quality of players & what the opponents give you - Importance of organization for counterpressing situations via more passes in the oppositions half

HP

398,772 次观看 • 2 年前

“if you could be any animal what would you choose?”

“if you could be any animal what would you choose?”

Kevin Sanji

24,749 次观看 • 1 年前

If you could be from another country what would it be

If you could be from another country what would it be

Mr Decent🦅

332,672 次观看 • 5 个月前

If you could get HP to autograph ONE thing, what would it be? 😅

If you could get HP to autograph ONE thing, what would it be? 😅

Mumbai Indians

112,951 次观看 • 3 个月前

Larry Ellison openly describes a plan to unify all citizen data via digital ID into a single national database—designed to be fully “consumable” by AI models. Centralized data + AI = enormous power.

Larry Ellison openly describes a plan to unify all citizen data via digital ID into a single national database—designed to be fully “consumable” by AI models. Centralized data + AI = enormous power.

Truthseeker

11,770 次观看 • 7 个月前

If you could give 1 piece of advice to men on how to be healthy, what would it be?

If you could give 1 piece of advice to men on how to be healthy, what would it be?

Bathmate

10,459 次观看 • 1 年前

🤔 If you could post one thing to our Instagram story, what would it be?

🤔 If you could post one thing to our Instagram story, what would it be?

Sunderland AFC

36,791 次观看 • 24 天前

if you could recommend one song to someone getting into kpop what would it be?

if you could recommend one song to someone getting into kpop what would it be?

nugu archive 📁

51,991 次观看 • 6 个月前

What if you could vote without leaving your house? Here's why Mobile Voting founder Bradley Tusk says creating a safe and secure way to vote via phone could be key to increasing voter turn out and reducing political extremism:

What if you could vote without leaving your house? Here's why Mobile Voting founder Bradley Tusk says creating a safe and secure way to vote via phone could be key to increasing voter turn out and reducing political extremism:

TED Talks

35,121 次观看 • 10 个月前

What happens when vision🤝 robotics meet? Happy to share our new work on Pretraining Robotic Foundational Models!🔥 ARM4R is an Autoregressive Robotic Model that leverages low-level 4D Representations learned from human video data to yield a better robotic model. Berkeley AI Research😊

What happens when vision🤝 robotics meet? Happy to share our new work on Pretraining Robotic Foundational Models!🔥 ARM4R is an Autoregressive Robotic Model that leverages low-level 4D Representations learned from human video data to yield a better robotic model. Berkeley AI Research😊

Roei Herzig

62,805 次观看 • 1 年前

Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a recipe for finetuning large vision-language-action (VLA) models with heterogeneous sensory data, such as vision, touch, sound, and more. Details in the thread 👇

Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a recipe for finetuning large vision-language-action (VLA) models with heterogeneous sensory data, such as vision, touch, sound, and more. Details in the thread 👇

Carlo Sferrazza

46,111 次观看 • 1 年前