I have no goal here. Just sharing my opinions. Not failing to do anything.

Yeah being aggressive is good for driving people away. And yknow given that your goal is actually to drive people away I was wrong to say it’s immature.

I just don’t like aggression. I don’t go on the internet looking for fights.

They can do whatever they want.

I don’t care what other people do, I just ignore people I don’t think are worth failing with.

And yeah pass judgement if you want, but how I choose to deal with people on the internet is up to me.

It’s just polarizing. You’re just making people more staunch in their beliefs or just annoying people who would rather not deal with aggression (like myself)

If your goal is to drive people away and make a space where everyone just agrees with you all the time then it’s effective.

Exactly

I’m dead 💀

Yknow I’m talking about on social media platforms, right?

Frothing at the mouth raging at someone on a social media platform doesn’t do anything but cause more radicalization, so I just ignore people instead. I don’t spend most of my life fighting with people on the internet over politics.

Indexing and tools like llamaindex use LLM generated embeddings to “intelligently” search for similar documents to a search query.

Those documents are usually fed into an LLM as part of the prompt (eg. context)

Hey yknow that’s a good point.

Yes, you can craft your prompt in such a way that if the llm doesn’t know about a referenced legal document it will ask for it, so you can then paste the relevant section of that document into the prompt to provide it with that information.

I’d encourage you to look up some info on prompting LLMs and LLM context.

They’re powerful tools, so it’s good to really learn how to use them, especially for important applications like legalese translators and rent negotiators.

Generally, training an llm is a bad way to provide it with information. “In-context learning” is probably what you’re looking for. Basically just pasting relevant info and documents into your prompt.

You might try fine tuning an existing model on a large dataset of legalese, but then it’ll be more likely to generate responses that sound like legalese, which defeats the purpose

TL;DR Use in context learning to provide information to an LLM Use training and fine tuning to change how the language the llm generates sounds.

You didn’t present any ideas or solutions to argue against. There’s no argument happening here.

Nor are there strawmen because there’s no argument being made.

You said that there’s generally a lack of imagination with regards to this stuff and I was just sharing my opinions as to why.

I think most people (correctly imo) don’t see how a large enough company can operate without some hierarchy, which seems to run up against the idea of being entirely equally employee owned.

There’s always going to be leaders (manager or just someone who others listen to) That person necessarily has more responsibility and control than his peers and is justly compensated more (otherwise nobody would put in extra work, say, to train as an engineer or doctor)

That person has their own interests that don’t always line up with the company and may use their influence to guide the company in a way that benefits them.

Suddenly you have a worker class and a bourgeois-esque class.

Most people (incorrectly imo) think that the “unbiased” checks and balances in government counteract that.

If there’s another option that accounts for hierarchies in large employee owned and operated companies let me know…. please

EDIT: large as in number of employees

It def adds some flavor to the social media political scene

The news bot posts are more annoying to me than the politics.

At least here there’s more than trump lovers and trump haters.

I hate scrolling into infinite bot posts with 0 comments

Looks like they got that number from this quote from another arstechnica article ”…OpenAI admitted that its AI Classifier was not “fully reliable,” correctly identifying only 26 percent of AI-written text as “likely AI-written” and incorrectly labeling human-written works 9 percent of the time”

Seems like it mostly wasn’t confident enough to make a judgement, but 26% it correctly detected ai text and 9% incorrectly identified human text as ai text. It doesn’t tell us how often it labeled AI text as human text or how often it was just unsure.

EDIT: this article https://arstechnica.com/information-technology/2023/07/openai-discontinues-its-ai-writing-detector-due-to-low-rate-of-accuracy/

lol

Awe well I still hope you have a good day.

Lol

And here we see an aggressive communist in the wild. You have a good day, dude

Ok, buddy