Replying to: https://x.com/DJiafei/status/1866594802126688484 Conservation of Information. True knowledge is measured by its impact and utility, not by sales If your Multimodal Language Models are not capable of interacting with all humans to improve spatial reasoning in humans – you have failed society, even if you get lots of investors and prizes from a few.
Read More »
Replying to https://x.com/marinkazitnik/status/1866666246726468080 All domain specific languages in all human languages can be losslessly simplified. Marinka Zitnik, It is not the AIs who struggle, it is the humans behind the green curtains (reference to wizard of oz) who try to make them look intelligent and witty and engaging to keep people coming back. But
Read More »
reply to https://x.com/curiouswavefn/status/1866536050090447050 Ilya Prigogine was only one of many. He did write about gravity. Nico Van Kampen When I was at UT Austin and hanging out with Prigogine’s group, there were many others around him who were as or more inciteful. I suggest you spend time and study what Nico Van Kampen was doing
Read More »
replying to https://x.com/docmilanfar/status/1866671417443619080 Your “noise” is always someone else’s precious and irreplaceable data – preserve it carefully. If you “remove noise” you must preserve the noise in a separate data stream. So when, later, as it will happen, you find your cleaning algorithm is way too simplistic, and essentially blocks all future super-resolution efforts —
Read More »
https://x.com/abhiramiyer/status/1866175915996434707 You need to stop listening to the people around you. They are making you look at too small pieces and narrow specialties. These notions are not wrong, but they have a global context and you are not making anything fundamentally new. These are easy digital and quantum functions. Copying the brain is like sketching
Read More »
https://x.com/keenanisalive/status/1866251675440460234 Many fairly random and chaotic looking things are often nothing surprising at all. Take any sequence and count unique values, the probability distribution ( as counts) tells you how many unique tokens are needed to store it losslessly. A log function fits for estimation, but so does the tail of a Poisson or normal
Read More »
I talked with OpenAI ChatGPT about the early universe. We are talking about Z=22 Richard Collins: Now if the early universe were dense it would form stars and black holes faster. With faster rates of accretion, it might well have fast creation of more black holes, more collisions and mergers of black holes. Also, if
Read More »
Issues in using AIs for “meningitis” and other exact terms on the Internet Richard Collins: What are the routes of infection for meningitis? Separate fungal, bacterial and viral. How good can diagnoses be if done by AIs with good tools and tests? The checklists with probabilities that are done by traditional databases and statistical frequencies
Read More »
Respectful, Responsible Humans and AIs: A New Approach to Handling Public Social Media Inputs I notice many X accounts of government agencies, cities, countries, large corporations, public entities, famous people, and groups say “This account is not monitored 24/7.” With the tools we have today, this doesn’t need to be the case. AI
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 »