Loading video...

Video Failed to Load

Go Home

Diffusion models leverage a variety of samplers. Deterministic methods like DDIM produce orderly paths. In contrast, stochastic samplers like DDPM produce chaotic trajectories. Despite their differences, both methods draw valid samples from the underlying distribution.

alec_helbling's profile picture

Alec Helbling

10,805 subscribers

101,989 views • 1 year ago •via X (Twitter)

Health & WellnessEducationScience & Technology
LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Stochastic and deterministic sampling strategies for diffusion models produce strikingly different trajectories, but both ultimately achieve the same aim. I had a great time presenting our work, Diffusion Explorer, this week at IEEE VIS in Vienna.
0:14

Stochastic and deterministic sampling strategies for diffusion models produce strikingly different trajectories, but both ultimately achieve the same aim. I had a great time presenting our work, Diffusion Explorer, this week at IEEE VIS in Vienna.

Alec Helbling

279,036 views • 7 months ago

Flow matching produces smooth, deterministic trajectories. In contrast, the sampling process of a diffusion model is chaotic, resembling the random motion of gas particles.
0:13

Flow matching produces smooth, deterministic trajectories. In contrast, the sampling process of a diffusion model is chaotic, resembling the random motion of gas particles.

Alec Helbling

428,834 views • 1 year ago

One common trick in MCMC methods is called 'burn-in'. It allows you to produce better samples by letting the chain first converge to the target distribution, discarding the first few samples which may be out of distribution.
0:16

One common trick in MCMC methods is called 'burn-in'. It allows you to produce better samples by letting the chain first converge to the target distribution, discarding the first few samples which may be out of distribution.

Alec Helbling

34,728 views • 1 year ago

Langevin Monte Carlo allows you to draw samples from a probability distribution using its log gradient ∇ log p(x). By performing a sort of gradient ascent with noise you can navigate around the distribution. Langevin MC is heavily related to modern diffusion models.
0:19

Langevin Monte Carlo allows you to draw samples from a probability distribution using its log gradient ∇ log p(x). By performing a sort of gradient ascent with noise you can navigate around the distribution. Langevin MC is heavily related to modern diffusion models.

Alec Helbling

96,842 views • 1 year ago

Israel employs inhumane and degrading methods in the distribution of humanitarian aid to Gaza's residents.
0:10

Israel employs inhumane and degrading methods in the distribution of humanitarian aid to Gaza's residents.

TIMES OF GAZA

18,096 views • 10 months ago

What $30 of fresh produce looks like in El Salvador 🇸🇻
0:15

What $30 of fresh produce looks like in El Salvador 🇸🇻

Salvadoran Pride

41,942 views • 1 year ago

I made a tool called Diffusion Explorer that lets you to train and visualize simple 2D diffusion and flow models live in the browser. You can draw your own distributions and observe how the generated samples converge during training. Try it live 👇
0:15

I made a tool called Diffusion Explorer that lets you to train and visualize simple 2D diffusion and flow models live in the browser. You can draw your own distributions and observe how the generated samples converge during training. Try it live 👇

Alec Helbling

73,103 views • 1 year ago

MIXERS AND HARMONIZERS TAMA MALI????? BIKINI’S A TOTAL BOP TALAGA!!! BINI FLAMES TRACK SAMPLERS #BINI_FLAMES #BINI
0:11

MIXERS AND HARMONIZERS TAMA MALI????? BIKINI’S A TOTAL BOP TALAGA!!! BINI FLAMES TRACK SAMPLERS #BINI_FLAMES #BINI

jam

68,193 views • 7 months ago

The latent space of earlier generative models like GANS can linearly encode concepts of the data. What if the data was model weights? We present weights2weights, a subspace in diffusion weights that behaves as an interpretable latent space over customized diffusion models.
0:16

The latent space of earlier generative models like GANS can linearly encode concepts of the data. What if the data was model weights? We present weights2weights, a subspace in diffusion weights that behaves as an interpretable latent space over customized diffusion models.

Amil Dravid

94,262 views • 2 years ago

Keeping it fresh for 2026 and beyond. Krown Produce and Aston Martin Aramco are excited to announce a multi-year deal, with the Canadian leader in premium produce and pioneer in importing for retail and wholesale distribution.
0:18

Keeping it fresh for 2026 and beyond. Krown Produce and Aston Martin Aramco are excited to announce a multi-year deal, with the Canadian leader in premium produce and pioneer in importing for retail and wholesale distribution.

Aston Martin Aramco F1 Team

32,196 views • 8 months ago

Google is making progress on their diffusion models... It's now as good as a Gemini 2.0 Flash Lite. The writing is on the wall, a majority of language AI use in the future will be diffusion models.
0:14

Google is making progress on their diffusion models... It's now as good as a Gemini 2.0 Flash Lite. The writing is on the wall, a majority of language AI use in the future will be diffusion models.

Carlos E. Perez

193,952 views • 5 months ago

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models "Block diffusion sequentially generates blocks of tokens by performing diffusion within each block and conditioning on previous blocks. By combining strength from autoregressive and diffusion models, block diffusion overcomes the limitations of both approaches by supporting variable-length, higher-quality generation and improving inference efficiency with KV caching and parallel sampling."
0:19

Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models "Block diffusion sequentially generates blocks of tokens by performing diffusion within each block and conditioning on previous blocks. By combining strength from autoregressive and diffusion models, block diffusion overcomes the limitations of both approaches by supporting variable-length, higher-quality generation and improving inference efficiency with KV caching and parallel sampling."

Tanishq Mathew Abraham, Ph.D.

21,813 views • 1 year ago

Capturing smoke from burning plastic in a bag demonstrates the significant amount of pollution a single fire can produce
0:10

Capturing smoke from burning plastic in a bag demonstrates the significant amount of pollution a single fire can produce

Interesting AF

355,793 views • 1 year ago

Cats have a wide spectrum of vocalizations, from the 'standard' meow, to purrs, to chirps and trills. Sometimes they produce these lower pitch vocalizations that vaguely sound like a human speech.
0:11

Cats have a wide spectrum of vocalizations, from the 'standard' meow, to purrs, to chirps and trills. Sometimes they produce these lower pitch vocalizations that vaguely sound like a human speech.

Massimo

26,606 views • 1 year ago

Printing using sand like this can also produce good products!!
0:16

Printing using sand like this can also produce good products!!

icel 🥟

67,300 views • 2 months ago

one of the tail brushing methods of all time
0:15

Sensitive content

one of the tail brushing methods of all time

emmpy

283,836 views • 10 months ago

Prop 50 would produce many bizarrely shaped districts, like the “elephants trunk” that would lump liberal Sacramento voters into a district primarily in the Sierra Nevada to rob conservatives of their voice in Congress. We’ll fight this corrupt scheme tooth and nail.
0:17

Prop 50 would produce many bizarrely shaped districts, like the “elephants trunk” that would lump liberal Sacramento voters into a district primarily in the Sierra Nevada to rob conservatives of their voice in Congress. We’ll fight this corrupt scheme tooth and nail.

Bruce Lou

52,115 views • 8 months ago

Sending Jerry from BIG JET TV wishes for a speedy recovery and hope he can soon produce great livestream moments like this.
0:14

Sending Jerry from BIG JET TV wishes for a speedy recovery and hope he can soon produce great livestream moments like this.

Francisco Cunha

80,049 views • 3 months ago