Hacker News Reader

Comment by Zetaphor | Hacker News Reader

What they did was immediately obvious and trivial to replicate

They make it clear that their edge is in their synthetic data. Train your own jevs miss the point. And if you have that much data, you didnt need jev or these replacements in the first place.

HN's desire to pretend the data pipeline doesn't exist or isnt meaningful is silly.

I'd argue that a tiny fine-tuned Jev on my dataset that I can run for practically free is far more valuable if the idea is to make this a dependency in real work

People have been making their own calibrated classifiers and decision makers for a very long time. Mine outperform jev. But that's not the point of jev.

What is...?

i guess it's not needing to finetune for each specific use case

That is indeed reasonable. But you still need a bunch of trusted data to validate the accuracy of Jev.

Yes. The difference is you can immediately benchmark and iterate via prompting without having to retrain. It's such a time saver.

Is it though? I have found it much easier to fit/diagnose/improve simple classifiers on embeddings/hash vectorized features/etc than iterate on prompts, especially true when using the more modern LLM's (like gpt 5.6 Luna) where you can't even set the temperature to get any sense of determinism.

I do note though, that is infinitely easier to 'deploy' a Jev/LLM based solution than a data/model pipeline.

The comparison i'm making is against a training of encoder/small decoder model so I think we agree on most things. I dont think jev replaces the benefit of embeddings nor think they are mutually exclusive. All part of a handy utility belt.

Then the probabilities are not calibrated. Jev is basically the LLM version of the "is-even" library.

Being ZERO-SHOT. No fine tuning or re-training needed. Why are people so dense over this model?

Reply on news.ycombinator.com