Hacker News

> difference between an instruct based re-ranker and laya/jev I just don't see it

Main difference is that laya/jev/et-al give you a zero-shot classifier that requires no training. You can prompt engineer your way to a quick fairly reliable cheap enough decision engine that you can use to iterate quickly (by prompt engineering).

Right now a lot of people are doing this with LLMs and it's too slow and expensive.

Imo the right iterative approach to productionizing these systems is something like:

    1. Build it with an LLM. Iterate on the prompt
    2. Start building a real-world dataset
    3. When the prompt works, turn it into a clear rubric for Jev or similar
    4. Keep iterating until desired accuracy achieved
    5. Use the real-world evals you've built to train a custom classifier fine-tuned to your needs

Jev's value becomes more apparent when the task is a moving target. eg an auto-mode classifier.

Calibrated probability across multi task with zero shot I guess. A reranker is single task and tuning it make it even more narrow. And I guess some piping to make multiclass efficient since you cannot mask logprob for independent questions in the same output space without throwing calibration away.

I have a lot of semi-practical examples of how you can use this model wrapped in unix-ish tools - https://github.com/aurorainfra/grev (readme links to docs of each tool with some more or less practical examples)

Really I think "smart grep" is a pretty good one ('look for an error looking vaguely like this'). Also I think sql-based shell history + decision model is quite good to make the last 'which one of those choices is best fit gives users past few commands' etc.

Is for when you want an AI to make a decision. If you have been using gpt or claude or open source models for that, than it’s a way cheaper alternative.

And if you have not been, it’s for when you have to extract the context from text. When you have numbers or fixed options, it’s just a matter of code.

So if you find yourself having to decide if a given user comment is a refund_request, that’s for that.

It’s not perfect, you still have to fine-tune (or calibrate) using examples you have (and keep those examples updated over time). But it’s way better than trying to parse text with regexes.

I've been following jevbench twice a day for the past week and that's been a lot of fun. Latest update:

Rank System Score Public / sealed accuracy Evidence

1 decider-4b v2 64.13 83.5% / 34.7% Evaluator-run, offline

2 Jev 1.13 63.29 86.6% / 36.7% Evaluator-run API

3 JevK5 v0.2 62.04 85.3% / 33.1% Evaluator-run

4 Cygnet 12B 61.76 87.9% / 33.8% Evaluator-run, offline

5 Hopper 59.43 82.3% / 34.1% Evaluator-run

28 Kev 4B 36.14 66.2% / 22.4% Evaluator-run

41 Laya 421M 30.25 58.4% / 30.8% Evaluator-run

https://benchmarkheaven.com/jev-models

Yes. JEV generalizes better because they probably have an enormous corpus and trained on it for a long time. Laya's out of the box model is much weaker. However, in the age of LLM's it's incredibly easy and cheap to generate large datasets to fine tune laya for your task, and the training loop is pretty quick and cheap too.

It's so easy that I question why I would ever pay for JEV when eventually I'll have done enough random things that I will also have a large corpus and likely a general model as well.

Developer here. You're right, Laya is a lot weaker than Jev, especially on harder queries. It's a small model, so it's fast, but that's the trade-off. The open models that get close to Jev are much bigger, and running those is what I'm working on next.

In my experience it's not close and the benchmarks I've seen don't reflect my experience at all.

But I'm guessing people will find the right training regime and data mix soon to close the gap.

But big things I see are instability and inaccuracy - like pick a random problem.

one day, perhaps people will click through to the laya author's arxiv paper content and the why may become clearer, you won't have to read it, a skim will suffice

Nothing yet. Unfortunately it sometimes feels like our industry has been overrun by grifters and chancers.

I’m sure this has been a gradual and long decline. Maybe it even started with the dot com boom and accelerated with crypto. With AI it seems to have got worse.

Because what they did is kinda trivial. Its basically like the Dropbox comment really[0], except here you don't need petabytes of storage and infinite VC pockets.

After chatgpt everything in AI mostly became LLMs and building wrappers around them. It's like people forgot how to do ML.

To those of us who actually trained models back in the day, its kind of cute to see people wowed by a classifier. Yes, this is 0 shot and doesn't need training (most people wanting this would've used structured output, this is cool because it's cheaper and faster). But anyone with basic ML knowledge could've built this in a few hours.

The question is mostly why wasn't this productized. And it's interesting indeed that it took this long to become a finished product.

[0] https://news.ycombinator.com/item?id=9224

Are you saying laya copied from jev, and released in two weeks? If so I don’t thinks it’s quite as simple a story as that. https://xtxinversexty.com/layas-prior-art-claim-is-absurd/

Presumably the training recipe and training dataset itself cannot be easily copied in a week or two. So if they want to shut down these competitor models they need to make it obvious how they are better than them.

"Decision model" is just marketing jargon.

decision model = classifier

system one model = small non-reasoning LLM

noul = boolean

confidence = f(probabilities)

It's sad to see how gullible engineers are today.

text classification is equivalente to decision. This is exactly the same thing Jev does.

Fair point, that example is basically classification. I'll change it to something that looks more like a real decision.

next vLLM release will have this

if you use gateways, GoModel support the S1 endpoints, my favorite feature is the virtual models, stable name, I can swap out the backing model(s)

https://gomodel.enterpilot.io/docs/getting-started/quickstar...

(the "kev" in the docs is my fault, I should have said Jev / System1 in my feature request)

Depends on the model. The small ones I support today are well below Jev on harder queries, but fine for simple, well-defined questions. The open models that get close to Jev are bigger, and I'm adding support for those next.

wait another week or so for vLLM's next release

there's this thing with a bunch of similar models https://huggingface.co/spaces/multimodalart/jev-decision-ind...

top open one is trained by perplexity cto for $3k, kinda cool https://x.com/denisyarats/status/2102252088067850507

The link rgbrgb posted is a good overview. The best open ones are close to Jev now, but they're big models. And I agree, if you have an eval set for a fixed task, a trained classifier is the better choice.

Fair, and I'd be happy if they did. Ollaya uses the same API as Jev, so your code isn't tied to it either way

and that ollama is go-llama and not rust, so it's not really the ollama of anything

open-source is the only way for safe AI development. whoever doesn’t share the weights/code will lag behind.

Thanks!

hey Claude, download and run vllm nightly for me

(already merged)

GoModel (gateway) already supports Jev like endpoints too

https://gomodel.enterpilot.io/docs/providers/jev