Ask HN: Is anyone doing anything cool with tiny language models? • Hacker News Board

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#311

Post by dutchbookmaker » Wed, Jan 22, 2025, 10:37 PM UTC

Etheryte wrote: post #198

kortilla wrote: post #167

Earlier quoted context omitted.

Right, but if 55% is avg over the last few years, “buy stock” is going to be correct more than not. https://www.crestmontresearch.com/docs/Stock-Yo-Yo.pdf

I think this is a good highlight of why context and reality checks are incredibly important when doing work like this. At first glance, it might look like 55% is a really good result, but in the previous year, a flat buy every day strategy would've been right 56.7% of the time.

55% means basically nothing in this context if even money. Long 45% to 55% is most likely completely random because it is symmetric with shorting 45% to 55%

Exactly what you would expect from a language model making random stock picks.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#312

Post by halJordan » Wed, Jan 22, 2025, 11:02 PM UTC

cwmoore wrote: post #63

I'm playing with the idea of identifying logical fallacies stated by live broadcasters.

Logical fallacies are oftentimes totally relevant during anything that is not predicate logic. I'm not wrong for saying "The Surgeon General says smoking is bad, you shouldn't smoke." That's a perfectly reasonable appeal to authority.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#313

Post by prettyblocks » Wed, Jan 22, 2025, 11:05 PM UTC

RhysU wrote: post #5

"Comedy Writing With Small Generative Models" by Jamie Brew (Strange Loop 2023) https://m.youtube.com/watch?v=M2o4f_2L0No Spend the 45 minutes watching this talk. It is a delight. If you are unsure, wait until the speaker picks up the guitar.

Excellent share - nice to see people doing cool things with the tech while not taking themselves too seriously.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#314

Post by thesz » Wed, Jan 22, 2025, 11:48 PM UTC

linsomniac wrote: post #128

I have this idea that a tiny LM would be good at canonicalizing entered real estate addresses. We currently buy a data set and software from Experian, but it feels like something an LM might be very good at. There are lots of weirdnesses in address entry that regexes have a hard time with. We know the bulk of addresses a user might be entering, unless it's a totally new property, so we should be able to train it on t…

From my experience (2018), run LLM output through beam search over different choices of canonicalization of certain part of text. Even 3-gram models (yeah, 2018) fare better this way.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#315

Post by genewitch » Thu, Jan 23, 2025, 12:29 AM UTC

halJordan wrote: post #312

cwmoore wrote: post #63

I'm playing with the idea of identifying logical fallacies stated by live broadcasters.

Logical fallacies are oftentimes totally relevant during anything that is not predicate logic. I'm not wrong for saying "The Surgeon General says smoking is bad, you shouldn't smoke." That's a perfectly reasonable appeal to authority.

It's still a fallacy, though. I hope we can agree on that part. If you have something map-reducing audio to timestamps of fallacies by who said them it makes it gamified and you can use the information shown to decide how much weight to give to their words.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#316

Post by sky2224 » Thu, Jan 23, 2025, 2:00 AM UTC

behohippy wrote: post #224

Dansvidania wrote: post #16

Earlier quoted context omitted.

this sounds pretty cool, do you have any video/media of it?

I don't have a video but here's a pic of the output: https://imgur.com/ip8GWIh

The next step is to format it so it looks like an endless starwars intro.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#317

Post by kortilla » Thu, Jan 23, 2025, 4:50 AM UTC

shiftingleft wrote: post #226

kortilla wrote: post #163

Earlier quoted context omitted.

Those don’t really work: https://jamanetwork.com/journals/jamanetworkopen/fullarticle...

Do they help deter people from becoming smokers in the first place?

Not sure if much serious research has been put into it. I would be suspicious of it deterring them because a lot of initial smoking happens in social situations where friends pass out individual cigarettes.

By the time someone buys their own pack they are probably hooked.

I suspect the obscene taxes blocking out young folks is one of the most effective strategies

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#318

Post by Mukina » Thu, Jan 23, 2025, 6:53 AM UTC

sidravi1 wrote: post #129

We fine-tuned a Gemma 2B to identify urgent messages sent by new and expecting mothers on a government-run maternal health helpline. https://idinsight.github.io/tech-blog/blog/enhancing_materna...

Super cool. What a simple and powerful way to help mothers in need. Thanks for sharing.

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#319

Post by swifthesitation » Thu, Jan 23, 2025, 7:18 AM UTC

whalesalad wrote: post #243

dh1011 wrote: post #130

I copied all the text from this post and used an LLM to generate a list of all the ideas. I do the same for other similar HN post .

chatgpt did a stellar job parsing the "books on hard things" thread from a little while ago. my prompt was: Can you identify all the books here, sorted by a weight which is determined based on a combo of the number of votes the comment has, the number of sub-comments, or the number of repeat mentions. Ideally retain hyperlinks if possible.

could you link the HN thread?

Re: Ask HN: Is anyone doing anything cool with tiny language models?

#320

Post by computers3333 » Thu, Jan 23, 2025, 7:56 AM UTC

tinco wrote: post #201

computers3333 wrote: post #176

https://gophersignal.com – I built GopherSignal! It's a lightweight tool that summarizes Hacker News articles. For example, here’s what it outputs for this very post, "Ask HN: Is anyone doing anything cool with tiny language models?": "A user inquires about the use of tiny language models for interesting applications, such as spam filtering and cookie notice detection. A developer shares their experience with using O…

That's cool, I really like it. One piece of feedback: I am usually more interested in the HN comments than in the original article. If you'd include a link to the comments then I might switch to GopherSignal as a replacement for the HN frontpage. My flow is generally: Look at the title and the amount of upvotes to decide if I'm interested in the article. Then view the comments to see if there's interesting discussion…

EDIT: Apologies for breaking things earlier while trying to fix it! I’ve been working on updating it and got the upvote count and comment link in there. Wondering what you think about these updates—appreciate any feedback! Thanks again for helping me improve it!

https://gophersignal.com