Richard Collins: I see many references to “diffusion models”. I know that chemical diffusion follows certain patterns and people do try to model it. This one is “a trained diffusion model” part of a Kihara Lab project to improve protein structure models with cryo electron microscope data. They call theirs “DiffModeler”. Is “diffusion modelling” a
Read More »
I am looking at solar flows , with chemical and isotopic compositions, densities and mass. I do not have a telescope but I wanted to share what I am doing with Helioviewer, aiming to add some features perhaps. If you are far ahead, please have patience. This is SDO AIA 171 with 15 minute running
Read More »
Yuntian Deng @yuntiandeng Interesting analysis by @NZZTech on WildChat, focusing on German-language conversations! They found that most German-language users treat ChatGPT like a search engine, with 45% of queries seeking concrete information. Grateful to see our work featured! x.com/NZZTech/status… NZZ Technologie @NZZTech Nov 29 Seit zwei Jahren gibt es Chat-GPT. Nun liefert ein Datensatz Einblick
Read More »
Conclusion: A World That Remembers Richard Collins: For things like Medicare health and drug plans it seems the confusion and complexity of the current system benefits the insurance companies, not the users. I see 70 plans offered and no tools to compare them. Just a long list of vague terms and no independent verification or
Read More »
Recommendations for Managing Large-Scale Parallel Development with AI – start with a formal feedback process based on real projects Richard Collins: That worked, sort of. You got the pixel counts to work but the RG RB GB scattergrams disappeared. The fundamental problem is your people do not know how to manage and
Read More »
Trying to get anything useful out of Grok – expect it will take them ten years to get off the ground Richard Collins: Is your interface improving so that you can save conversations? X Grok : Yes, my interface is indeed improving to better save conversations. ChatGPT, for instance, now automatically saves your conversations within
Read More »
Artificial intelligence today relies on massive data labeling, coding, curation, constant monitoring and fine tuning, no customer service yet, nor social responsibility or liability. Behind the scenes are legions of human laborers – shoveling coal into yet another “new” engine of progress. What do you think? Richard Collins, The Internet Foundation
Solar Data, global data, all AIs, all humans, Dana and Brian Trilogy, a future with eternal intelligences [ Please be aware that I had to make fairly strong statements to evoke consistent responses. Read the responses as a rough framework of terms and directions in society. It gets a bit gung-ho. I have had hundreds
Read More »
https://www.facebook.com/groups/1278689162297491 Could you explain what is going on in AutoStakkert when it is finding 10% of 5000 frames? I have a vague idea, but you are using it. What does it emphasize and what do you think is happening there? I do not have a telescope. I use data from the Internet. So I never
Read More »
Science Advances @ScienceAdvances Researchers discover that using foldable bottlebrush polymers as network strands provides a strategy to decouple the tradeoffs between stiffness and extensibility. My Comment: If these are made for “permanent” structural colors, might they be more dangerous than dyes? And harder to remove from humans and living things? I am asking you because
Read More »