MoreThanText/morethantext
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Update Home

jbaskin 2025-12-11 13:55:10 +00:00
parent ff6e9653ba
commit 091c1a1858
1 changed files with 9 additions and 2 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
Home.md

@ -24,6 +24,13 @@ There is a problem with trustworthiness right now. One of the ways of addressin
## Automatic Language Translation
Not everyone speaks and reads English. The application should respond to the user in their native language. Yes this does mean the use of Large Language Models (LLM), which most people refer to as AI; however, it should only be used as a place holder until a human can do a proper translation.
Not everyone speaks and reads English. The application should respond to the user in thpeeir native language. Yes this does mean the use of Large Language Models (LLM), which most people refer to as AI; however, it should only be used as a place holder until a human can do a proper translation.
LLMs are used for pattern recognition. That means that it finds information like this list of words in English match to this list of words in Japanese 95% of the time. An LLM translator recommends the list of words that match the highest percentage. That is why LLMs miss context clues at best, or produce incorrect garbage at worst. The latter is commonly referred to as hallucinating. Since the application will primarily be dealing with technical information with limited context clues, I believe that LLMs should be useful. We will have to see as the system develops and gains users.
LLMs are used for pattern recognition. That means that it finds information like this list of words in English match to this list of words in Japanese 95% of the time. An LLM translator recommends the list of words that match the highest percentage. That is why LLMs miss context clues at best, or produce incorrect garbage at worst. The latter is commonly referred to as hallucinating. Since the application will primarily be dealing with technical information with limited context clues, I believe that LLMs should be useful. We will have to see as the system develops and gains users.
<h2>SVG circle Element</h2>
<svg height="100" width="100" xmlns="http://www.w3.org/2000/svg">
<circle r="45" cx="50" cy="50" fill="red" />
Sorry, your browser does not support inline SVG.
</svg>