Hi @peihaowang 🤗
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/2603.09221.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models for instance), you can also claim the paper as yours which will show up on your public profile at HF, add Github and project page URLs.
I saw you've released the core implementation for the TTC-Net on GitHub—the hardware-efficient LQR solver approach is very interesting! In the paper, you mentioned integrating TTC layers as adapters into pretrained LLMs to achieve significant gains on MATH-500 and AIME.
Would you like to host these pre-trained adapter checkpoints on https://huggingface.co/models?
Hosting them on Hugging Face will provide better visibility and discoverability for your work. We can add specific tags (like mathematical reasoning or optimal control) so that researchers can easily find and use the adapters.
If you're interested, I'm leaving a guide here. For adapter-based models, people can easily download and apply them using the huggingface_hub library or the PyTorchModelHubMixin class.
After they are uploaded, we can link the checkpoints directly to the paper page so people can discover your artifacts while reading about your method.
Let me know if you're interested or if you need any guidance!
Kind regards,
Niels
Hi @peihaowang 🤗
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/2603.09221.
The paper page lets people discuss about your paper and lets them find artifacts about it (your models for instance), you can also claim the paper as yours which will show up on your public profile at HF, add Github and project page URLs.
I saw you've released the core implementation for the TTC-Net on GitHub—the hardware-efficient LQR solver approach is very interesting! In the paper, you mentioned integrating TTC layers as adapters into pretrained LLMs to achieve significant gains on MATH-500 and AIME.
Would you like to host these pre-trained adapter checkpoints on https://huggingface.co/models?
Hosting them on Hugging Face will provide better visibility and discoverability for your work. We can add specific tags (like mathematical reasoning or optimal control) so that researchers can easily find and use the adapters.
If you're interested, I'm leaving a guide here. For adapter-based models, people can easily download and apply them using the
huggingface_hublibrary or the PyTorchModelHubMixin class.After they are uploaded, we can link the checkpoints directly to the paper page so people can discover your artifacts while reading about your method.
Let me know if you're interested or if you need any guidance!
Kind regards,
Niels