Flying Scholars – CS

135 posts

Flying Scholars – CS banner
Flying Scholars – CS

Flying Scholars – CS

@FlyScholarsCS

Reviewing Entity. Smuggler of ideas. Imagining the future of Peer Review. FAQ: https://t.co/kGNULyLV9C review template: https://t.co/6ktN8p6Q9d

Inscrit le Mayıs 2021
410 Abonnements132 Abonnés
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
@flowersslop I think they meant "log progress": double the dataset size, get a (small) constant increase in performance.
English
0
2
2
48
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
> me waiting for one of these companies to release a simple search engine on anna's archive, instead of slop-washing it through "AI"
AI BS Detector tweet media
English
0
1
1
321
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
american dystopia
AI BS Detector tweet mediaAI BS Detector tweet media
CY
0
1
0
167
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
Question for a mathematician: if the fourier transform of a Dirac at 0 is the sum of all 0-phased fourier components, what is the fourier transform of a signal containing 2 diracs, one at 0 and one at 1?
English
1
1
1
56
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
7/7 papers accepted at #IAMTHEBEST. Not one worth mentioning the content. Congratulations in the replies will be met with condescension. GFY
English
0
1
0
78
Flying Scholars – CS retweeté
latenightneuro
latenightneuro@latenightneuro·
"The Neuron as a Direct Data-Driven Controller" 🔗 arxiv.org/abs/2401.01489 🎯 Thinking of biological neurons as trying to stabilize their own local environment could explain many of their properties: spike-timing dependent plasticity, contrast adaptation, temporal integration..
latenightneuro tweet medialatenightneuro tweet medialatenightneuro tweet medialatenightneuro tweet media
English
1
1
2
134
Flying Scholars – CS retweeté
AI BS Detector
AI BS Detector@AI_BS_Detector·
Calling bullshit.
Lior Alexander@LiorOnAI

This is big. Researchers were able to generate images directly from brain EEG signals. It's the best performing “thoughts-to-image" model yet to be released. Here's how they did it: ▸ EEG signals are collected from the brain and a masked signal pre-training is used to create an effective EEG encoder. ▸ The encoder is fine-tuned with limited EEG-image pairs using Stable Diffusion. ▸ The EEG, text, and image spaces are aligned using CLIP encoders to create a unified space. ▸The EEG embedding representation is optimized for image generation using a loss function. ▸ High-quality images are generated from the optimized EEG representation using Stable Diffusion.

English
0
1
1
64
Flying Scholars – CS
Flying Scholars – CS@FlyScholarsCS·
5/ Users should be able to post on different channels they create, and their followers should be able to select the channels they wish to follow
English
1
0
0
0
Flying Scholars – CS
Flying Scholars – CS@FlyScholarsCS·
✨Dreaming up my perfect twitter✨ 1/ Likes should *not* be stochastic retweets. They should be a private mark of appreciation.
English
1
0
0
0
Tony Zador
Tony Zador@TonyZador·
@neuralreckoning I still believe that we need a mech for tagging interesting papers, ie postpub review not just for truth but also for interest. I just think it should be decoupled from gatekeeping. I am a fan of @ylecun's "reviewing entities"--like postpub journals yann.lecun.com/ex/pamphlets/p…
English
3
3
20
0
Dan Goodman
Dan Goodman@neuralreckoning·
This has led to some quite strongly worded disagreement, so perhaps it's worth expanding a bit on this to explain myself and for me to understand the disagreement better.
Dan Goodman@neuralreckoning

It might seem hypocritical to not review papers but expect others to review mine. But I'm not asking you to do differently from me. I encourage you to join me in rejecting this system and refusing to do pre-publication review entirely. We change things by forcing a painful crisis

English
8
4
39
0
Flying Scholars – CS
Flying Scholars – CS@FlyScholarsCS·
@recursus Hey 🫵, you seem tired of the hype; how about sharing your thoughts in a public review?
English
0
0
1
0
Daniel Bear
Daniel Bear@recursus·
AFAICT there's no result in the paper that implies it's solved the issues with counting. Fig A.15 and A.16 indicate that Imagen is slightly better than DALL-E2 and GLIDE, but don't see any evaluation of absolute accuracy.
English
1
0
2
0
Flying Scholars – CS
Flying Scholars – CS@FlyScholarsCS·
@3scorciav @SattlerTorsten @ftmguney @eccvconf @CSProfKGD Hi Victor, what we are really missing at this point is researchers that would be willing to allocate their time and talent to write quality public reviews for our service. Understandibly, there is little incentive to spend time on this task for researchers at the moment.
English
1
0
2
0
F. Güney
F. Güney@ftm_guney·
I'm probably the most annoying AC ever 🤓 - Which parts of the method do you think need more details? - What kind of ablation studies/datasets specifically do you recommend? - Could you add some references to verify your claim on "limited novelty" and point to the similarities?
English
6
4
98
0

Découvrir

@flowersslop @JamesKilner @TonyZador @ylecun @neuralreckoning @recursus @3scorciav @SattlerTorsten