Everyone now talks about the architecture that's not auto regressive and does lightning fast probability prediction with a json schema. I worked on this literally one year back in March 2025, published an arxiv paper, pushed the model to huggingface along with the pypi package and training dataset. And then one year later, a frontier lab came, proposing the same idea like literal breakthrough without technical papers, open weights and no open dataset. For anyones information the main guiding model is RL not embedding model or LLM
Also the second work published in September 2025 was exactly the same one jev proposed now
Paper: https://arxiv.org/abs/2510.01237
My model uses PPO over sequence embeddings to output turn-by-turn conversion trajectories (probabilities from 0.0 to 1.0).
Jev uses parallel sampling (trained via RLCD) to output confidence distributions and schema choices.
It's incredibly frustrating that the thing that you made with months of hard work, sweat and sleepless night is architecturally similar with the vertical use case and don't get the support you deserve because frontier lab build something horizontal. The open-source story in general
The world's heavily about marketing, resources, connections, and signaling, unfortunately. You probably needed to market it in a bigger forum with shinier claims to attract attention (I don't think a paper on Arxiv is enough).
Pity because without the frustration it would’ve been a good post.
Catch is you’ve got intuition, but looking around instead of forward.
Do it again, open-source it, either you’d quietly bring down few companies, or, when you’re close, you’d get offers from them.
The who’s done what is a dying paradigm.
From what I can tell these are classifiers trained for a single task. What excites people about Jev is that it can do zero-shot structured responses for arbitrary prompts. Now, this isn't new either; models like GLiNER have been around for a while. But Jev appears significantly more flexible and polished while still being cheap and fast.
Paper: https://arxiv.org/abs/2503.23303
Model: https://huggingface.co/DeepMostInnovations/sales-conversion-model-reinf-learning
Dataset: https://huggingface.co/datasets/DeepMostInnovations/saas-sales-conversations
Also the second work published in September 2025 was exactly the same one jev proposed now
Paper: https://arxiv.org/abs/2510.01237
My model uses PPO over sequence embeddings to output turn-by-turn conversion trajectories (probabilities from 0.0 to 1.0).
Jev uses parallel sampling (trained via RLCD) to output confidence distributions and schema choices.
It's incredibly frustrating that the thing that you made with months of hard work, sweat and sleepless night is architecturally similar with the vertical use case and don't get the support you deserve because frontier lab build something horizontal. The open-source story in general
Feeling that open-source community do not get the recognition they deserve
https://github.com/vllm-project/vllm/pull/57250
The world's heavily about marketing, resources, connections, and signaling, unfortunately. You probably needed to market it in a bigger forum with shinier claims to attract attention (I don't think a paper on Arxiv is enough).
Put it on your CV
"I can come up with revolutionary ideas one year earlier than entire billion dollar organizations"
"I inventet Jev a year earlier"
terrible, antisocial and wrong take
Pity because without the frustration it would’ve been a good post. Catch is you’ve got intuition, but looking around instead of forward. Do it again, open-source it, either you’d quietly bring down few companies, or, when you’re close, you’d get offers from them. The who’s done what is a dying paradigm.
We are human , frustration is something we feel when same kinda architecture is closed sourced and we celebrate it
Absolutely. Besides, when we see injustice we shouldn't be silent about it because that would encourage more of it.
I saw the post about Jev and the lack of basic information felt wrong and bizarre, I now see the most likely reason for it.
The best thing to do now, is to develop your open source project further and beat the Jev-ers, that story is very motivational and I'm interested.
Can you (or someone) provide some context for this post? What's Jev? What's the innovation here that was duplicated by the frontier lab?
From what I can tell these are classifiers trained for a single task. What excites people about Jev is that it can do zero-shot structured responses for arbitrary prompts. Now, this isn't new either; models like GLiNER have been around for a while. But Jev appears significantly more flexible and polished while still being cheap and fast.