Daksh Aggarwal (@dakshces) - Twitter Profili | Zamantika Mersobahis Locabet

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·5 Mar

Ever wish you could tell a video model what to achieve, rather than just how to move? Introducing our CVPR 2026 paper, Goal Force! Instead of simulating a direct push, our model plans the entire causal chain (the "how") to achieve your specified goal (the "what"). 🧵(1/n)

English

3

15

64

9.1K

Daksh Aggarwal retweetledi

Chen Sun@jesu9·28 May

Can you inject causal control to models trained solely on video data? How about eliciting their understanding of physical forces when animating a photo? Nate's latest project shows that both are promising directions to explore! I tried it on my favorite photo, and you should too! github.com/brown-palm/for…

Nate Gillman@GillmanLab

Ever wish you could turn your video generator into a controllable physics simulator? We're thrilled to introduce Force Prompting! Animate any image with physical forces and get fine-grained control, without needing any physics simulator or 3D assets at inference. 🧵(1/n)

English

2

5

51

6K

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·28 May

Ever wish you could turn your video generator into a controllable physics simulator? We're thrilled to introduce Force Prompting! Animate any image with physical forces and get fine-grained control, without needing any physics simulator or 3D assets at inference. 🧵(1/n)

English

8

66

318

44.3K

Daksh Aggarwal retweetledi

François Chollet@fchollet·3 Kas

People who treat prediction markets as some kind of magical oracle are hopelessly lost. Even the valuation of very high-liquidity public stocks is anything but efficient. Now, election bets? Forget about it.

English

19

7

206

20.2K

Daksh Aggarwal retweetledi

François Chollet@fchollet·3 Kas

Remember that prediction markets are very low-liquidity -- with a couple M$ you can clear the entire order book. They do *not* encode any kind of "efficient" prediction of future outcomes given all available data. They encode the uninformed opinion of a handful of rich gamblers.

The Kobeissi Letter@KobeissiLetter

BREAKING: Prediction markets now show Kamala Harris with a 4 percentage point higher likelihood of winning the 2024 election over Donald Trump. Just days ago, Donald Trump held a 30 percentage point lead on Harris, according to @Kalshi. Wow.

English

28

51

655

72.4K

Daksh Aggarwal retweetledi

Chen Sun@jesu9·31 Eki

.@GillmanLab and the team have done it again! 🎉 Fourier Head brings theory-driven enhancements to LLMs when modeling continuous tokens! The best part? It actually works and is easy to implement!

Nate Gillman@GillmanLab

LLMs are powerful sequence modeling tools! They not only can generate language, but also actions for playing video games, or numerical values for forecasting time series. Can we help LLMs better model these continuous "tokens"? Our answer: Fourier series! Let me explain… 🧵(1/n)

English

0

2

18

3.1K

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·31 Eki

Check out our project page for more details (paper, code, visuals…) nategillman.com/fourier-head This was a team effort :) joint work with @dakshces @mik3fr33man @saurasingh @jesu9, a fun collaboration between @BrownUniversity / @GoogleDeepMind, and of course, Fourier's head (n/n)

English

2

9

91

6.1K

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·31 Eki

LLMs are powerful sequence modeling tools! They not only can generate language, but also actions for playing video games, or numerical values for forecasting time series. Can we help LLMs better model these continuous "tokens"? Our answer: Fourier series! Let me explain… 🧵(1/n)

English

25

210

1.6K

206.7K

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·23 Tem

Check out our @ICMLconf poster tomorrow if you want to learn how you can improve human motion generation performance by injecting expert physics knowledge during training :) presenting w/ @jesu9 + @dakshces I'll be in Vienna through Sunday!! #ICML2024 nategillman.com/sc-sc.html

English

1

15

2.4K

Daksh Aggarwal retweetledi

Chen Sun@jesu9·3 May

Now accepted by #ICML2024, code is available at github.com/nate-gillman/s… !

Nate Gillman@GillmanLab

Excited to share our latest preprint: “Self-Correcting Self-Consuming Loops for Generative Model Training”. It's a step towards generative AI models that can learn from the universe of data they generate!! 🤖(1/n)

English

1

5

31

4.1K

Daksh Aggarwal retweetledi

Chen Sun@jesu9·21 Şub

Video genmos (e.g. #Sora) are touted as "world simulators". Can we leverage their outputs to create more powerful genmos? Our paper arxiv.org/abs/2402.07087 demonstrates mathematically that a naive solution would lead to model collapse, but a self-correction op would fix that!

Nate Gillman@GillmanLab

Excited to share our latest preprint: “Self-Correcting Self-Consuming Loops for Generative Model Training”. It's a step towards generative AI models that can learn from the universe of data they generate!! 🤖(1/n)

English

1

14

77

14.2K

Daksh Aggarwal retweetledi

Nate Gillman@GillmanLab·21 Şub

Excited to share our latest preprint: “Self-Correcting Self-Consuming Loops for Generative Model Training”. It's a step towards generative AI models that can learn from the universe of data they generate!! 🤖(1/n)

English

4

20

109

34.8K

Daksh Aggarwal

Keşfet