Moving Symbol sets, global open methods

Eric Brachmann @eric_brachmann Map-relative Pose Regression CVPR2024 highlight)

For years absolute pose regression did not work. There was some success by massively synthesising scene-specific data. We train scene-agnostic APR and it works.

Paper: https://arxiv.org/abs/2404.09884
Page: https://nianticlabs.github.io/marepo https://pic.twitter.com/uLNMFMQ42L
Replying to @eric_brachmann

There ought to be a concise moving symbol set to describe precisely this algorithm, not force humans to look at example images, read words that are not in the languages and experience of most humans, and not really give complete links to the whole of what you did and who makes it possible. Better than words, better than images, better than symbols on paper. I am fairly certain it is possible now. If rich monopolies will pay back what society has granted them, rather than taking advantage. If geographic entities will help be fair to all 8.1 Billion humans. If many humans who are suffering will not give up hope.
Β 

Richard Collins, The Internet Foundation

 


About 12 hours ago, we introduced a bug in our web server, which resulted in Multimer results being visualized as Monomers. If you used it during this time frame, please try it again. It is fixed just now. I am sorry for the inconvience. x.com/thesteinegger/…
Martin Steinegger πŸ‡ΊπŸ‡¦
@thesteinegger
Β·
Apr 15
Foldseek-Multimer is a protein complex aligner that is up to 10,000x times faster than SOTA methods without sacrificing quality, enabling the comparison of billions of complex pairs per day. 1/5
πŸ“„ https://biorxiv.org/content/10.1101/2024.04.14.589414v1
πŸ’Ύ https://github.com/steineggerlab/foldseek
🌐 https://search.foldseek.com https://pic.twitter.com/AewkD59vPq
Replying to @thesteinegger

 

You are showing what you have to do now to make this happen. Not what interfaces should be when there are global open methods that all humans can see, access, use, understand and use for communication and discovery. Please put more effort into making your tools universally accessible to all humans (and their infant AIs), so your methods are not in isolation for a few thousands, but open and useful immediately to all 8.1 Billion humans, and one day their true AIs.
Β 
(no trustworthy AIs yet. Groups still refuse to use permanent global open resource methods).
Β 
Richard Collins, The Internet Foundation
Richard K Collins

About: Richard K Collins

Director, The Internet Foundation Studying formation and optimized collaboration of global communities. Applying the Internet to solve global problems and build sustainable communities. Internet policies, standards and best practices.


Leave a Reply

Your email address will not be published. Required fields are marked *