51–60 of 79 posts
Earlier quoted context omitted.
And yet, it's so much easier to deploy an LLM, either through a service or on prem.
It's easier to do a lot of things. That doesn't make it better.
I wonder if using prompts to get the sentiment in LLM is enough? So we do not need to do any fine-tuning anymore?
I also think it *could* be less of a problem than you might think. If we treat the scale as arbitrary (which I think is a safe thing to do), then movement along the scale could be sufficient to ascertain *something*
what's up with the title flips from > 350M Tokens Don't Lie: Love And Hate In Hacker News, to > LLM-based sentiment analysis of Hacker News posts, to > LLM-based sentiment analysis of Hacker News posts between Jan 2020 and June 2023
Earlier quoted context omitted.
Because LLMs WILL dominate all NLP use cases, whether you like it or not. Its like the linux of operating systems. Sure you can handwrite up some custom OS more specialized for a purpose. But its much easier to just use linux, which everyone understands on a basic level and is extremely robust, and modifying it slightly for the end goal. And saying "Traditional sentiment analysis" tools are "Battle tested" is laughab…
Aside from a half dozen or so zeros, you're right on.
Earlier quoted context omitted.
I'm wondering how their LLM parsing 250 mil words in 9 hours compares with performance of traditional sentiment analysis. Also, many exisiting sentiment analysis tools have a lot of research behind them that can be referenced when interpreting the results (known confounds etc). I don't think there is yet an equivalent for the LLM approach
And yet, it's so much easier to deploy an LLM, either through a service or on prem.
Earlier quoted context omitted.
And yet, it's so much easier to deploy an LLM, either through a service or on prem.
What do you mean? Deploying something like spaCy is far easier than deploying an LLM in my experience.
LLMs really are almost magic in how they can help in this space; and setting them up is often just getting an API key and throwing some money and webservice calls at them.
Why is everything only plotted between 4 and 8 if the scale of the least liked topic should be 0 and most liked should be 9. Also 4.5 is the midpoint, but 4 is displayed as bright red and 6 is a muted gray blue, why? This makes no sense except to be psychologically disingenuous. And no 5s? What is even going on in that LLM?
It's nice to see this scale used outside of The Good Place.
> Hate : Torture Great work folks, glad we can all agree on that one. Interesting that they used an LLM for this. I mean it makes sense and the data seems to pass the pub test but I, in my ignorance, would not have assumed that a language model would be well suited for number crunching.
Seems we mostly agree on hating Atlassian, too, so it's working as intended.