Release IB-TPO optimized models on Hugging Face

Hi @bowiehsu 🤗

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/2605.28109.

The paper page lets people discuss your paper and find related artifacts (like your models). You can also claim the paper as yours, which will show up on your public profile at HF, and add your GitHub and project page URLs.

I saw in the paper that you developed IB-TPO and evaluated it on various Qwen3 models. I noticed that the GitHub link provided (`https://github.com/alibaba/EfficientRL`) currently leads to a 404 page. Are you planning to release the code and the fine-tuned checkpoints (1.7B, 8B, 14B) soon?

Would you like to host those models on https://huggingface.co/models? Hosting on Hugging Face will give your work more visibility and enable better discoverability through metadata tags. We can also link the models directly to the paper page so researchers can find and use them easily.

If you're down, I'm leaving a guide [here](https://huggingface.co/docs/hub/models-uploading). If you are using a custom setup, you can use the [PyTorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) class which adds `from_pretrained` and `push_to_hub` to your model, making it very easy for others to download and use.

Let me know if you're interested or need any guidance!

Kind regards,

Niels

Start with the Hugging Face model-uploading guide and the PyTorchModelHubMixin documentation linked in the issue. Confirm the IB-TPO artifacts and the 1.7B, 8B, and 14B fine-tuned checkpoints are available, then make the models discoverable and link them to the paper page.

Written by the indexing model from the issue text.