Shulou(Shulou.com)11/24 Report--

The latest version of Mathematica Wolfram officially introduces the large language Model (LLM).

This means that now all it takes to do scientific calculation is to "say".

For example, if you want to make a picture, just say:

Draw a red and a green semicircle.

Draw a red and green semicircle.

Based on this, more requirements can be made in natural language:

I want filled semicircles next to each other. And no axes.

I want a filled semicircle adjacent to each other. There is no axis.

Continue to make changes in natural language:

Make these so the flat sides form a vertical diameter.

Let these flat sides form a vertical diameter.

Finally, to put the two semicircles together, just say:

Make these part of the same circle, without a space in between.

Connect the two parts into a circle with no gap in the middle.

It is not difficult to see that it is quite convenient to use natural language to do scientific calculation.

In the official words of Mathematica Wolfram, it is:

Even if you do not have a priori knowledge and do not know the Wolfram language, you can easily handle scientific calculations.

When netizens saw this new feature, they said:

Buy whatever you want!

We can also correct errors by ourselves. Let's use another example to experience the Mathematica that introduces LLM:

Make a table of the first 20 twin primes.

Make a table listing the first 20 double primes.

Similarly, based on the description of the natural language, Mathematica started his work "conveniently".

In the meantime, however, Mathematica discovered that his code was written incorrectly, and began to admit the error and correct it:

I'm sorry. I seem to overestimate the biprime of the first 50 primes. Let's adjust the code to find the first 20 biprime numbers with no predefined limits.

Even if it doesn't write a piece of code itself, it's very good at explaining what happens when an error is generated:

Officials say a variety of new structures have been added to the newly introduced LLM,Wolfram language.

For example, in Chat Notebook, chat units (chat cell) and chat blocks (chatblock) are set up.

In order to facilitate the use of LLM, officials have also set up a "Wolfram prompt repository" (Wolfram Prompt Repository).

The prompt repository consists of two main categories:

The first is personalization, which is used for "styling" and otherwise notifies chat interactions.

The second is function hints and modifier hints.

Function hints are designed to get LLM to do something specific, such as summing up a paragraph of text, or suggesting a joke (not very good at it at the moment).

Modifier hints are used to determine how LLM should modify its output, such as translating it into different human languages or keeping it at a certain length.

The introduction of more updates to LLM is just one of the 13.3 updates.

For example, the newly integrated "line integral" and "surface integral" allow users to integrate scalar and vector functions on lines and surfaces.

Officially, this feature opens up new possibilities for mathematical computing.

In addition, version 13.3 has been updated in more functions and visualization.

For details, you can poke the link below.

Reference link:

[1] https://writings.stephenwolfram.com/2023/06/llm-tech-and-a-lot-more-version-13-3-of-wolfram-language-and-mathematica/

[2] https://twitter.com/stephen_wolfram/status/1674120239435456514

[3] https://news.ycombinator.com/item?id=36529610

This article is from the official account of Wechat: quantum bit (ID:QbitAI), author: Jin Lei

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.