starchild3001

11 ms·

91.
by starchild3001 2y ago
I pretty much agree with the perspective in this paper. That is, depending on your definition, either 1) AGI has already happened (in a narrow domain of question answering, machines can answer any question & converse, nearly as well as
92.
by starchild3001 2y ago
He's only 50% evil :)
93.
by starchild3001 2y ago
Not a major problem in my experience if you use the aistudio version of the model. Claude tends to be the most conservative in my experience. Gemini 2.0 and 4o appear similar.
94.
by starchild3001 2y ago
Yes.
95.
by starchild3001 2y ago
I use all top of the line models everyday. Not for coding, but for general "cognitive" tasks like research, thinking, analysis, writing etc. What Google calls Gemini Pro 2.0 has been my most favorite model for the past couple of m
96.
by starchild3001 2y ago
I'm well aware of that. That is not what openai calls "deep research".
97.
by starchild3001 2y ago
Not really accurate. The "Search" functionality you're describing in DeepSeek is comparable to OpenAI's existing "Search GPT." OpenAI's recent announcement refers to a more advanced capability, similar to
98.
by starchild3001 2y ago
What a poorly informed article. It's very shallow and out of touch with LLM research. As it stands 6-12 months old models are system 1 thinkers, everybody knows this and knew this even at the time. You need system 2 thinking (test time
99.
by starchild3001 2y ago
Background in telecommunications/signal processing (phd) + applied math. I studied finance + machine learning for two years as a hobby before the switch. In the evenings and weekends. Studied data structures + algorithms + python for t
100.
by starchild3001 2y ago
I graduated from undergrad 25 yrs ago. I worked in 5 different business sectors with highly overlapping but also quite distinct skill sets (4.5 yrs of academia/writing papers, 5 yrs of telecom/signal processing engineering, 7 yrs
101.
by starchild3001 2y ago
yes, it turned out to be a scam to conflate "diversity in thought" with skin color or gender. The term "diversity" became too toxic. The term "equity" was morally dubious / wrong to begin with. Inclusivity
102.
by starchild3001 2y ago
Guys, it's a major 21st century skill to learn how to use LLMs. In fact, it's probably the biggest skill anyone can develop today. So please be a responsible driver, learn how to use LLMs. Here's one way to get the most milea
103.
by starchild3001 2y ago
I think the jump from ChatGPT to GPT4 was also something like 130 ELO points. This roughly equals 2/3 preference for the new model, 1/3 preference for the old model. That's roughly how much top models have improved for an &qu
104.
by starchild3001 2y ago
In 2024, AI made astonishing progress across the board. Top models have improved by roughly 130 ELO points on public leaderboards, while costs for running those same models have dropped many fold. Smaller, low-latency models now offer near–
105.
by starchild3001 2y ago
A visit to Google scholar can answer your question. Having made deeper and broader contributions to the field is usually how researchers become more famous. E.g. most PhD advisors are more renowned than their students because they've m
106.
by starchild3001 2y ago
Intelligence comes in many forms and flavors. ARC prize questions are just one version of it -- perhaps measuring more human-like pattern recognition than true intelligence. Can machines be more human-like in their pattern recognition? O3 m
107.
by starchild3001 2y ago
Thank you for being sensible We have so many "intelligence professors", who can conveniently dismiss SoTA AIs with a sleight of hand (Cholet, LeCunn etc), yet completely ignore the superintelligent traits these AIs already exhibi
108.
by starchild3001 2y ago
Tell me how Search AI Overviews aren't using Gemini?
109.
by starchild3001 2y ago
I just tried it after your message. It has like 5-10 voices. Many of them very reasonable, some of them quite good on my pixel phone.
110.
by starchild3001 3y ago
That's a gross mischaracterization. There's no stalking. It's your data kept in a server that no human eyes can see, and algorithmically matched with the best ads possible. That's the kind of ads I know, served respectfu
111.
by starchild3001 3y ago
Hey, current Googler here. I joined in 2017. For those who are looking from outside to those who reminisce past Google, I have an update for you: Google is still a f'ing amazing workplace. It's easily the best company I worked for
112.
by starchild3001 3y ago
LOL, this is the stupidest list I've seen in a long time. Tech sector being very dynamic and full of alternative opportunities clearly creates the short tenure over there. Whereas relatively stagnant industries with fewer job opportuni
113.
by starchild3001 3y ago
A Second LLM's take on this lawsuit can be found below. I'd love to see OpenAI address these complaints publicly and without incurring any additional damages to NYT. The document is a legal complaint filed by The New York Times Co
114.
by starchild3001 3y ago
I asked an LLM to summarize the 69 page lawsuit. It does a decent job. Didn't infringe on any copyrights in the process :) Here is a summary of the key points from the legal complaint filed by The New York Times against Microsoft and O
115.
by starchild3001 3y ago
WeWork's business model had concerning parallels to the risks that led to the 2008 financial crisis. Banks then made risky bets chasing short-term profits while accumulating long-term vulnerabilities. Similarly, WeWork took on expensiv
116.
by starchild3001 3y ago
Those who lived through past market busts will remember enron, worldcom, tyco etc. (2000s), lehman, bear sterns, aig etc (2008s). This time around we have Theranos, FTX, WeWork,... Different names, same story. Market excesses, almost always
117.
by starchild3001 3y ago
College admissions are broken, plain and simple. Completely opaque. Success criteria unclear, appears highly subjective, subject to gaming and monetary influence, officially biased against non-alumni & non-affiliated children. Simply F&
118.
by starchild3001 3y ago
Given the animosity Elon Musk has displayed towards former and current Xitter employees, it should surprise us more that people keep working with him. People leaving him is not surprising. Deciding to work with him is surprising.
119.
by starchild3001 7y ago
Yes,I get the criticism against OOP. I kinda don't understand how FP solves the problem of shared mutable state... Or what alternatives there are for myraid design patterns. I coded in Scala for <1 yr. Loved its syntax and code clea
120.
by starchild3001 8y ago
I remember Google and Facebook valuations called a bubble many times around the time of their IPOs. The truth: We never know (guess what? They weren't).

More ›