Arthur Caetano

PhD Student @UCSB | HCI, XR, Hand-Object Interaction

[Atieh Taheri @atiehtaheri]

Feeling grateful and exhilarated after a phenomenal #UIST2023 finale! Honored with the People's Choice Best Poster Award—a dream come true. A huge thank you to everyone who supported and believed in our work. Your votes of confidence are truly inspiring! 🌟 #HardWorkPaysOff

[Atieh Taheri @atiehtaheri]

Had a blast presenting our Virtual Buddy poster @ #UIST2023 w @caetanolab!🌟 Missed our poster yesterday? No worries! Catch us @ 10:20am today to dive into our innovative work fostering inclusivity.🕥 And don't forget, your vote counts! Cast yours here: https://t.co/TV2g8906Hl 🗳️

[Bart Wronski 🇺🇦🇵🇸 @BartWronsk]

Some papers are so elegant they seem “obvious” (but are novel). https://t.co/H89KCyMQjb
Instead of implementing full differentiable renderer (tons of work and challenges), why not just render normally and then soft-splat the pixels into differentiable points? Much easier/faster!

[Arthur Caetano @caetanolab]

@MoritzW42 @generalcatalyst Looks great and the search bar is even better now! 🤩

[Julian Togelius @togelius]

Excited about this new work from our lab, where we show that neural cellular automata (essentially iteration over convnets) can be trained to do really interesting level generation in several domains. Not only that, we can use Map-Elites do get sets of diverse level generators!

[Pablo Samuel Castro @pcastr]

PSA: MuJoCo is offering free licenses for everyone until Oct. 31, 2021:

https://t.co/VjCaBHi8N6

[128 MHz @128_mhz]

Storing objects in your pocket freeform #gamedev #indiegame

[Andreas Köpf @neurosp1ke]

Join us tonight for the @ykilcher paper discussion:

`The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers`

September 4, 2021
18:00 to 20:00 UTC

Paper: https://t.co/rOCtSVbtVh

Yannic's Discord: https://t.co/UibUKqmk2g

[AK @_akhaliq]

3DStyleNet: Creating 3D Shapes with Geometric and Texture Style Variations
pdf: https://t.co/YTg3BfUvjh
abs: https://t.co/1rYQQKF26w
project page: https://t.co/Lqlt87YWOw

[MetallCore999 @MetallCore999]

Water shader + GPU ripples + Halfwater effect + Water physics 🌊
Many thanks who helped me figure out some things: @_staggart_ @pensionerov
And of course I couldn't have done it without a miracle as ASE! 😍
#madewithunity #indiedev #unity3d #AmplifyShaderEditor

[kaleidoscube ⌛ RECUR @kaleidoscube]

Since some of you are quite interested in some behind the scenes stuff, we made a very quick How-To-Stylized-Fire in @UnrealEngine4⁠.
All the fire-effects in our game use this setup as a base and it really helps us to create deep atmospheres!⁠

#indiedev #gamedev #ue4

[Bianca Pereira @BiaResearcher]

#AcademicTwitter what is the best feature of your reference management (Zotero, Mendeley, etc)?

best = you use most or has the highest impact on your work

Trying to decide (if) which I should use.

@PhDVoice @PostdocVoice @PhDinTech #AcademicChatter

[Marc Pollefeys @mapo1]

Check out our new dataset for visual localization https://t.co/igkDdG2d8E

[Kevin Patrick Murphy @sirbayes]

I am pleased to announce that the camera ready version of my new textbook, "Probabilistic Machine Learning: An Introduction", is finally available from https://t.co/dSlKkwYpLr. Hardcopies will be available from MIT Press in Feb 2022.

[ACM SIGGRAPH @siggraph]

We showcased some knock-out content this year at #SIGGRAPH2021, including "Control Strategies for Physically Simulated Characters Performing Two-player Competitive Sports" 🥊 Watch @twominutepapers' highlight plus conference attendees can watch the on-demand session until 29 Oct.

[Dr Adeayo Sotayo @DrSotayo]

@ClayGemma I'll give three useful sources:

• Academic Phrasebank - https://t.co/8Czuff2VvY

• Check out Book #8 in this list - https://t.co/gNzLXfZEV6

• Grammarly (https://t.co/I0MIg4rZ8A) or ProWritingAid for concise writing.

Good luck Gemma

[Pierre Vandermaesen @pyairvander]

Made a full capacity matrix out of tinyfolks #pixelart #1bit #gamedev #gameart

[Arthur Caetano @caetanolab]

https://t.co/JeFeDApFxB

[UCSB AI @UCSB_AI]

1/3 GPT-2 and similar #NLProc models are known to output memorized training data. Can they be prompted to output conspiracy theories? Yes.
@sharonlevy21's #acl2021nlp findings paper cleverly answers this question without access to the training dataset.
https://t.co/ZHvdNxgj81