Experiments to make GPT4 compensate for context

• I saw a high context conversation, so I tried to see if GPT4 could compensate for the context.
• It didn’t work very well, so a human supplemented and generated it again.
• Subsequent additions by the individuals themselves - Good abstraction and bad abstraction - Assumption quantity and decision making

shogochiai If you don’t solidify the basic fundamentals, the scaffolding, it will shift. It is always shifting. If you are too quick-witted, you can pile up assumptions and go far, far away.

shogochiai As with the Mandala PowerPoint presentation at Kasumigaseki, the story of a big circle of people who make multiple assumptions is the driving force behind the economy, so I don’t deny it in general. However, it is also a manner of betting, so we need to start from the footing without getting drunk. In fact, I don’t think there are many people who feel that they have made progress because they have been spinning around with their feet still in the air.

_tea_two “A is A” may be one of the cooking tools that can cook “abstraction” well.

Supplement the context and explain.

This Twitter post discusses the importance of assumptions and abstractions, but also expresses the opinion that those methods can lead in the wrong direction if they are not based on a solid foundation.

• The first tweet… In particular, he warns that people who are quick thinkers can build up their own assumptions quickly, but if those assumptions are not based on a solid foundation, they may reach erroneous conclusions.
• Second tweet… He specifically mentions the accumulation of assumptions in economics and policy.
  • PowerPoint presentation on Kasumigaseki (the center of Japanese government) (Mandala PowerPoint presentation on Kasumigaseki)
    • Stories that build up a large number of assumptions and reach a large conclusion (big circle) cannot generally be denied, as they are the driving force that keeps the economy moving.
    • However, it is like a bet (the manner of the betting booth)
    • I think a lot of people don’t get the sense that they are actually making progress if they ignore the basics as they go along.
• Last tweet… “A is A” is a tool that handles abstraction well, he said.
  • This underscores the importance of understanding the nature and basic properties of things, and then making abstractions based on that understanding.
  • In other words, the idea is that by understanding things and then creating abstract concepts based on them, we can understand them more deeply.

Referring to the following context

• The “scaffolding” here is a real phenomenon
• People who excel at studying for exams are skilled at building on linguistically given assumptions and constraints without questioning them.
  • There are many excellent people in area in Tokyo, but it is excellence based on this thought pattern
• A is A” is the subtitle of the third part of “Atlas Shrugs. Everyone must always see for themselves. Do you know what this novel is about? If you know, supplement it with context.

- The Shrugging Atlas is a philosophical novel written by [[Ayn Rand.]] that argues for [[self-centeredness]], rational [[selfishness]], and the importance of being free from group and other people's expectations. - "A is A" is the core philosophy of this work and represents the importance of understanding the nature of things and acting on that basis. - Each person is expected to see and judge things from his or her own perspective and act accordingly.

• Taking this context into account, the discussion in the original tweet can be said to be about the balance between real phenomena (“scaffolding”) and the assumptions and abstractions that are built upon them (“building”).

• The “brilliant” are described as having the ability to build upon given assumptions and constraints without questioning them.

  • This may indicate the ability to problem solve within a particular framework or paradigm.
  • However, if the framework itself is based on faulty assumptions, then solutions based on it may also be faulty.

• Kasumigaseki (the center of Japanese government) attracts the best and the brightest

  • but are the thinking patterns and frameworks they are using appropriate?
  • The question is also posed as to whether they accurately reflect real phenomena (“scaffolding”).

• The concept of “A is A” is cited in this discussion to emphasize the importance of understanding the nature of things and acting on that basis. In other words, it illustrates the importance of accurately understanding real phenomena and making assumptions and abstractions based on that understanding.

• Given the above context, this discussion prompts thought on the issue of standards of excellence and success, and how they should be based on an understanding of reality.

Understanding the Essence” is a little different.

• So piling one’s own thoughts on top of the thoughts of others is “gambling” and “making the stakes bigger” because it is merely piling hypotheses on top of hypotheses. Participants making their bets bigger in a gambling hall is the same as participants making investments in the economy, which makes the economy bigger and raises tax revenues. It is the same structure as when a betting house makes money through temple coins. Unlike mere gambling, it is much better because it increases the probability of success.
• On the other hand, in the shrugging atlas, the person who actually makes a new motor or new steel that works is lifted up. Even if a hypothesis is built up and a “plausible claim” is made, if actual experiments do not produce the expected results, then the “actual experimental results” are more correct than human thought. This empirical science pattern of thinking and behavior has created a community of successful people, where the golden rule is “you have to see for yourself.

- Thank you very much for the more specific explanation, I understand more clearly the meaning of "A is A" here and the reference to the "shrugging atlas". I also understand that "scaffold" refers to a real phenomenon, i.e., real data or results. - Based on this, the interpretation of each tweet is adjusted. - The first tweet: emphasizes the importance of having the basic understanding of reality ("scaffolding") necessary to understand and interpret things. It implies that we should base our understanding on real world data and results, not just words. - Second tweet: drawing big conclusions (the big circle) based on a multitude of assumptions and speculations is part of economic activity, and like investing and gambling, there are risks and rewards. However, he points out that if those assumptions are not based on real data and results, they may spin out of control and fail to provide a sense of progress. - A final tweet: drawing from the "Shrugging Atlas" the action guideline of "see for yourself," which emphasizes the importance of understanding and acting on the basis of reality. In other words, the message is that it is important to see reality for oneself and act accordingly, without relying on the thoughts and assumptions of others.

Given these contexts, these tweets can be interpreted as emphasizing the importance of thinking and acting based on real phenomena, or “scaffolding”. And they argue that its importance is particularly important in policy making, economic activities, and scientific problem solving.

---

This page is auto-translated from /nishio/GPT4に文脈を補わせる実験 using DeepL. If you looks something interesting but the auto-translated English is not good enough to understand it, feel free to let me know at @nishio_en. I’m very happy to spread my thought to non-Japanese readers.