• 0 Posts
  • 217 Comments
Joined 2 years ago
cake
Cake day: June 23rd, 2023

help-circle



  • Well, one line of my comment was about her being part of the CIA torture apparatus. The rest detailed how she, personally and with the help of the Democratic Party, did fascist things like subvert the will of the people by preventing real political choice.

    This isn’t a situation where we can pick the lesser of two evils and hold our nose. I’m outright saying she’s part of the problem — so fuck her and her insistence that we focus on only the kings they want us to be concerned about. And fuck her desire to step away from our moral duty to protect others from discrimination.

    I will not abandon my principles for word games.




  • You say “Not even close.” in response to the suggestion that Apple’s research can be used to improve benchmarks for AI performance, but then later say the article talks about how we might need different approaches to achieve reasoning.

    Now, mind you - achieving reasoning can only happen if the model is accurate and works well. And to have a good model, you must have good benchmarks.

    Not to belabor the point, but here’s what the article and study says:

    The article talks at length about the reliance on a standardized set of questions - GSM8K, and how the questions themselves may have made their way into the training data. It notes that modifying the questions dynamically leads to decreases in performance of the tested models, even if the complexity of the problem to be solved has not gone up.

    The third sentence of the paper (Abstract section) says this “While the performance of LLMs on GSM8K has significantly improved in recent years, it remains unclear whether their mathematical reasoning capabilities have genuinely advanced, raising questions about the reliability of the reported metrics.” The rest of the abstract goes on to discuss (paraphrased in layman’s terms) that LLM’s are ‘studying for the test’ and not generally achieving real reasoning capabilities.

    By presenting their methodology - dynamically changing the evaluation criteria to reduce data pollution and require models be capable of eliminating red herrings - the Apple researchers are offering a possible way benchmarking can be improved.
    Which is what the person you replied to stated.

    The commenter is fairly close, it seems.





  • I read a comment a few weeks ago from someone that is from around that area. Between J6 and the inauguration, apparently a bunch of concrete barriers sprung up and there were military personnel stationed throughout checkpoints at the Capitol complex.
    Biden is many things, but he and his cronies are not intentionally keeping security weak in hopes that a coup will happen, nor will he or the other two people that can activate them hesitate to deploy the National Guard.

    J6 was not a generalized failure of government processes or a mistake from lack of planning. There are standing plans, materiel, and personnel in place. The state of the Capitol and its defenses were a choice — an intentional failure, to open the doors for a coup.










  • I honestly can’t recall if it was some sort of geopolitical analysis in the comments or actual news anymore, but years ago I read that climate change and the collapse of the North Atlantic Current would eventually open up vast areas of Siberia to mining/drilling, improve farming conditions in Russia, harm farming, solar, and wind in Western Europe, while dropping the temps in Western Europe. It would also raise temps in the eastern/southern U.S. and make hurricanes more dangerous and economically damaging along the entire Atlantic and Gulf coasts.
    What I read concluded that climate change would be a major boon to Russia and any sensible leader there would want to facilitate it.