COP-GEN: Latent Diffusion Transformer for Copernicus Earth Observation Data
Paper • 2603.03239 • Published • 2
AI & ML interests
Earth Observation Datasets
Recent Activity
mespinosami
authored a
paper 3 days ago
Add task category and link to EarthEmbeddingExplorer paper
1
#3 opened 21 days ago
by 
nielsr
mikonvergence
posted an update 7 days ago
Post
153
🥕 Introducing BetaEarth - your own Earth embedding emulator [𝐏𝐫𝐞-𝐑𝐞𝐥𝐞𝐚𝐬𝐞]
The past year has brought many notable embedding products, like AlphaEarth, TESSERA or OlmoEarth. We are entering a phase where embeddings begin to act as a substitute for real observation data.
BetaEarth is an attempt to explore how much one can learn from a model based on its embeddings alone, and whether those embeddings can serve as a useful training target for other models. Huge credit to the AlphaEarth team for releasing the embedding archive openly — it's what made this kind of community-built extension possible.
[𝐁𝐞𝐭𝐚𝐄𝐚𝐫𝐭𝐡 𝐢𝐬 𝐧𝐨𝐭 𝐚 𝐟𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥 𝐛𝐮𝐭 𝐢𝐭 𝐭𝐫𝐢𝐞𝐬 𝐢𝐭𝐬 𝐛𝐞𝐬𝐭]
BetaEarth is a flexible (and relatively lightweight) emulator of the AlphaEarth annual product. It doesn't reproduce AlphaEarth's exact outputs, nor the product, but it reaches ~0.87 cosine similarity on held-out data and retains 97% of downstream land-cover classification accuracy. It only took 1-2 days to train.
It can encode any combination (including multi-temporal) of:
- Sentinel-2 L1C
- Sentinel-2 L2A
- Sentinel-1 RTC
- COP-DEM 30 product
The model weights are open, just like its training data (built exclusively using Major TOM). The GitHub repository provides a script for automated generation of embeddings across any footprint.
You can also try the workflow over small bounding boxes on the free Hugging Face web app!
⚙️ GitHub: https://github.com/asterisk-labs/beta-earth
🖥️ Web App: asterisk-labs/betaearth
🏭 Models: https://huggingface.co/collections/asterisk-labs/beta-earth
🟨 Colab: https://colab.research.google.com/github/asterisk-labs/beta-earth/blob/main/examples/generate_demo.ipynb
🗞️ Pre-print: https://github.com/asterisk-labs/beta-earth/blob/main/docs/beta_earth_preprint.pdf
The past year has brought many notable embedding products, like AlphaEarth, TESSERA or OlmoEarth. We are entering a phase where embeddings begin to act as a substitute for real observation data.
BetaEarth is an attempt to explore how much one can learn from a model based on its embeddings alone, and whether those embeddings can serve as a useful training target for other models. Huge credit to the AlphaEarth team for releasing the embedding archive openly — it's what made this kind of community-built extension possible.
[𝐁𝐞𝐭𝐚𝐄𝐚𝐫𝐭𝐡 𝐢𝐬 𝐧𝐨𝐭 𝐚 𝐟𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐦𝐨𝐝𝐞𝐥 𝐛𝐮𝐭 𝐢𝐭 𝐭𝐫𝐢𝐞𝐬 𝐢𝐭𝐬 𝐛𝐞𝐬𝐭]
BetaEarth is a flexible (and relatively lightweight) emulator of the AlphaEarth annual product. It doesn't reproduce AlphaEarth's exact outputs, nor the product, but it reaches ~0.87 cosine similarity on held-out data and retains 97% of downstream land-cover classification accuracy. It only took 1-2 days to train.
It can encode any combination (including multi-temporal) of:
- Sentinel-2 L1C
- Sentinel-2 L2A
- Sentinel-1 RTC
- COP-DEM 30 product
The model weights are open, just like its training data (built exclusively using Major TOM). The GitHub repository provides a script for automated generation of embeddings across any footprint.
You can also try the workflow over small bounding boxes on the free Hugging Face web app!
⚙️ GitHub: https://github.com/asterisk-labs/beta-earth
🖥️ Web App: asterisk-labs/betaearth
🏭 Models: https://huggingface.co/collections/asterisk-labs/beta-earth
🟨 Colab: https://colab.research.google.com/github/asterisk-labs/beta-earth/blob/main/examples/generate_demo.ipynb
🗞️ Pre-print: https://github.com/asterisk-labs/beta-earth/blob/main/docs/beta_earth_preprint.pdf
mikonvergence
updated a
dataset 12 days ago
mikonvergence
updated a
dataset 21 days ago
mikonvergence
published a
dataset 21 days ago
Smith42
authored 2
papers 21 days ago
mikonvergence
authored 2
papers 28 days ago
mikonvergence
updated a
dataset 28 days ago
Update dataset card with task category, paper links, and citation
1
#3 opened 30 days ago
by nielsr
blumenstiel
authored 2
papers about 2 months ago
Post
6820
We should really have a release date range slider on the /models page. Tired of "trending/most downloaded" being the best way to sort and still seeing models from 2023 on the first page just because they're embedded in enterprise pipelines and get downloaded repeatedly. "Recently Created/Recently Updated" don't solve the discovery problem considering the amount of noise to sift through.
Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.
Slight caveat: Trending actually does have some recency bias, but it's not strong/precise enough.
- 3 replies
mkluczek
updated a
dataset about 2 months ago
mkluczek
published a
dataset about 2 months ago
simidjievskin
authored 3
papers 2 months ago
How Well Does Your Tabular Generator Learn the Structure of Tabular Data?
Paper • 2503.09453 • Published
TabEBM: A Tabular Data Augmentation Method with Distinct Class-Specific Energy-Based Models
Paper • 2409.16118 • Published
ProtoGate: Prototype-based Neural Networks with Global-to-local Feature Selection for Tabular Biomedical Data
Paper • 2306.12330 • Published