Ethan

@torchcompiled

trying to feel the magic. cofounder at @leonardoai_ directing research at @canva

tampa - sydney

ethansmith2000.com

Joined April 2022

12KPosts 9KFollowers 845Following

You might like

@devdef

@KaliYuga_ai

@pharmapsychotic

@Shade_9S0

@multimodalart

@huemin_art

@takyon236

@thibaudz

@NerdyRodent

@proximasan

@sureailabs

@nin_artificial

@zippy731

@AiJoe_eth

@HvnsLstAngel

Pinned

Ethan

@torchcompiled

Feb 14

personally I feel like the inflection point was early 2022. The sweet spot where clip-guided diffusion was just taking off, forcing unconditional models to be conditional through strange patchwork of CLIP evaluating slices of the canvas at a time. It was like improv, always…

torchcompiled's tweet image. personally I feel like the inflection point was early 2022. The sweet spot where clip-guided diffusion was just taking off, forcing unconditional models to be conditional through strange patchwork of CLIP evaluating slices of the canvas at a time. It was like improv, always…

EPROM

@eprombeats

Feb 14

Image synthesis used to look so good. These are from 2021. I feel like this was an inflection point, and the space has metastasized into something abhorrent today (Grok, etc). Even with no legible representational forms, there was so much possibility in these images.

eprombeats's tweet image. Image synthesis used to look so good. These are from 2021. I feel like this was an inflection point, and the space has metastasized into something abhorrent today (Grok, etc). Even with no legible representational forms, there was so much possibility in these images.

Ethan reposted

Ethan

@torchcompiled

12 h

New post! As opposed to building reward models over human ratings and using them for RL, can a model develop its own reward function? Humans seem to develop their own aesthetic preferences through exploration and socializing. How can we mimic this for generative models?

torchcompiled's tweet image. New post! As opposed to building reward models over human ratings and using them for RL, can a model develop its own reward function?

Humans seem to develop their own aesthetic preferences through exploration and socializing.

How can we mimic this for generative models?

Ethan

@torchcompiled

9 h

2 raised solutions here: 1 captures the social aspect while leaving out the difficulty of exploration: That one's taste develops by learning about other's tastes, we could imagine training a generative model trained over a dataset of many reward models, and sample new plausible…

torchcompiled's tweet image. 2 raised solutions here:

1 captures the social aspect while leaving out the difficulty of exploration: That one's taste develops by learning about other's tastes, we could imagine training a generative model trained over a dataset of many reward models, and sample new plausible…

Ethan

@torchcompiled

12 h

Ethan reposted

Ethan

@torchcompiled

Sep 14

New post! I believe we can think of ourselves in two different lenses: an exact point of experience and the history of our patterns of behavior. Though the two are deeply interconnected.

torchcompiled's tweet image. New post! I believe we can think of ourselves in two different lenses: an exact point of experience and the history of our patterns of behavior. Though the two are deeply interconnected.

Ethan reposted

Ethan

@torchcompiled

Sep 16

Last one in this series! Can we pinpoint where the self is located in the mind? We know that a number of parts of the brain can be ablated with minimal personality changes while others are more damaging. Not to mention how the body at large plays a role.

torchcompiled's tweet image. Last one in this series! Can we pinpoint where the self is located in the mind? We know that a number of parts of the brain can be ablated with minimal personality changes while others are more damaging. Not to mention how the body at large plays a role.

Ethan reposted

Ethan

@torchcompiled

Sep 12

Yet another post on identity. Similar to the last, our model of the world is shaped by our observations of the world, which in turn precipitates our identities.

torchcompiled's tweet image. Yet another post on identity. Similar to the last, our model of the world is shaped by our observations of the world, which in turn precipitates our identities.

KaliYuga

@KaliYuga_ai

AK

@_akhaliq

hardmaru

@hardmaru

proxima centauri b

@proximasan

sure, ai

@sureailabs

TomLikesRobots🤖

@TomLikesRobots

Claire Silver 🌸

@ClaireSilver12

Roope Rainisto

@rainisto

illustrata

@illustrata_ai

pharmapsychotic

@pharmapsychotic

Guy Parsons

@GuyP

huemin

@huemin_art

Jeremy Torman ♏

@TormanJeremy

Dreaming Tulpa 🥓👑

@dreamingtulpa

AmliArt

@amli_art

BLΛC

@blac_ai

makeitrad

@makeitrad1

Emm | scenario.com

@emmanuel_2m

ErotemeArt.eth

@ErotemeArt

Thibaud Zamora

@thibaudz

akash

@akashtattva

fourth

@fourthfought

Manuel Garcia

@ManuelOrManny

diana gaia

@diah_gay

Grace Li

@grx_xce

learning

@reprompting

Vixen

@VividAura12

Chapy

@ChapyCode

Christian Hinrichsen

@DataHiresJP

Aditya Ramesh

@model_mechanic

Soran Ghaderi

@soranghadri

Le Duy Duc

@duyducle20

Kushpreet Singh

@kushpreets57

Kazz

@kazzorr_

Doraking

@doraking_en

jangmeongwoen

@pollawon2

Shivank Nigam

@NigamShivank

Fredrik K. Gustafsson

@fregu856

Angad Ahuja

@angadahujasin

zhiqiang xu

@xu_zhiqiang

박지호

@qkrwlgh0314

john

@john56391960913

Thanh Pham

@pth_opt

chintak

@chintak

$et_tu_deux's profile picture. Diffraction limit ignorant, applied category theoretician and chief metanarrative author #27 #lgrw #goblue #floodthezone Erev Shabbat! GPGPU to the car!$