Hi @XiangboGaoBarry 馃
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as yours got featured: https://huggingface.co/papers/2604.16272.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models and dataset for instance), and I see you've already shared the VEFX-Bench dataset and the 4B reward model on the Hub, which is awesome!
Would you like to also host the VEFX-Reward-32B model you've mentioned on https://huggingface.co/models?
I noticed in the GitHub README that it is currently listed as "coming soon". Hosting it on Hugging Face will give your work more visibility and enable better discoverability. We can add tags in the model cards so that people find the models easier, link it to the paper page, etc.
Additionally, we can link your current VEFX-Reward-4B and VEFX-Bench dataset to the paper page (read here) so people can easily find all your evaluation tools in one place.
If you're down for the 32B release, you can use the same process as the 4B model. If it's a custom PyTorch model, you can use the PyTorchModelHubMixin class which adds from_pretrained and push_to_hub to the model.
Let me know if you're interested or if you'll need any guidance when the 32B version is ready for release :)
Kind regards,
Niels
ML Engineer @ HF 馃
Hi @XiangboGaoBarry 馃
I'm Niels and work as part of the open-source team at Hugging Face. I discovered your work through Hugging Face's daily papers as yours got featured: https://huggingface.co/papers/2604.16272.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models and dataset for instance), and I see you've already shared the VEFX-Bench dataset and the 4B reward model on the Hub, which is awesome!
Would you like to also host the VEFX-Reward-32B model you've mentioned on https://huggingface.co/models?
I noticed in the GitHub README that it is currently listed as "coming soon". Hosting it on Hugging Face will give your work more visibility and enable better discoverability. We can add tags in the model cards so that people find the models easier, link it to the paper page, etc.
Additionally, we can link your current
VEFX-Reward-4BandVEFX-Benchdataset to the paper page (read here) so people can easily find all your evaluation tools in one place.If you're down for the 32B release, you can use the same process as the 4B model. If it's a custom PyTorch model, you can use the PyTorchModelHubMixin class which adds
from_pretrainedandpush_to_hubto the model.Let me know if you're interested or if you'll need any guidance when the 32B version is ready for release :)
Kind regards,
Niels
ML Engineer @ HF 馃