Hi @mikonvergence 🤗
Niels here from the open-source team at Hugging Face. I discovered your new work on EarthEmbeddingExplorer through Arxiv! It's great to see the Major TOM ecosystem expanding with such practical retrieval tools.
The paper page at hf.co/papers/2603.29441 lets people discuss your work and find related artifacts. You can also claim the paper to show it on your public profile and link your GitHub repository.
I noticed that while you've already hosted the DINOv2 and SigLIP embeddings for Major TOM on the Hub (which is awesome!), the paper also mentions providing ready-to-use embeddings for FarSLIP and SatCLIP. Would you like to host those on Hugging Face as well?
Hosting them under the Major-TOM organization will significantly improve their discoverability and allow researchers to easily load them using the datasets library:
from datasets import load_dataset
dataset = load_dataset("Major-TOM/Core-S2-FarSLIP")Additionally, since you've built the interface with Gradio and hosted it on ModelScope, it would be fantastic to also host a version of the EarthEmbeddingExplorer app on Hugging Face Spaces. We can provide you a ZeroGPU grant if the retrieval backend requires A100 GPUs for inference.
Let me know if you're interested or need any help with the uploads!
Cheers,
Niels
ML Engineer @ HF 🤗
Hi @mikonvergence 🤗
Niels here from the open-source team at Hugging Face. I discovered your new work on EarthEmbeddingExplorer through Arxiv! It's great to see the Major TOM ecosystem expanding with such practical retrieval tools.
The paper page at hf.co/papers/2603.29441 lets people discuss your work and find related artifacts. You can also claim the paper to show it on your public profile and link your GitHub repository.
I noticed that while you've already hosted the DINOv2 and SigLIP embeddings for Major TOM on the Hub (which is awesome!), the paper also mentions providing ready-to-use embeddings for FarSLIP and SatCLIP. Would you like to host those on Hugging Face as well?
Hosting them under the Major-TOM organization will significantly improve their discoverability and allow researchers to easily load them using the
datasetslibrary:Additionally, since you've built the interface with Gradio and hosted it on ModelScope, it would be fantastic to also host a version of the EarthEmbeddingExplorer app on Hugging Face Spaces. We can provide you a ZeroGPU grant if the retrieval backend requires A100 GPUs for inference.
Let me know if you're interested or need any help with the uploads!
Cheers,
Niels
ML Engineer @ HF 🤗