Research map / Vision-language and multimodal / Human and embodied modeling
CMB-ML: A Cosmic Microwave Background Dataset for the Oldest Possible Computer Vision Task
James Amato, Yunan Xie, Leonel Medina-Varela, Ammar Aljerwi, Adam McCutcheon, T. Seth Rippentrop, Kristian Gonzalez, Jacques Delabrouille, Mustapha Ishak, Nicholas Ruozzi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use cmbNNCS instead of CNILC | CMB-ML baselines · CMB-ML test, deconvolved | RMSE: 76.45 -> 23.26 | Improved |
| use cmbNNCS instead of CNILC | CMB-ML baselines · CMB-ML test, deconvolved | MAE: 59.83 -> 18.50 | Improved |
| use cmbNNCS instead of CNILC | CMB-ML baselines · CMB-ML test, 20.6 arcmin beam | RMSE: 8.686 -> 4.235 | Improved |
| use cmbNNCS instead of CNILC | CMB-ML baselines · CMB-ML test, 60 arcmin beam | RMSE: 5.887 -> 0.788 | Improved |
| use cmbNNCS instead of CNILC | CMB-ML baselines · CMB-ML test, deconvolved | PSNR: 33.29 -> 32.72 | No clear effect |
| smooth maps with 60 arcmin beam instead of 20.6 arcmin | CNILC · CMB-ML test | RMSE: 8.686 -> 5.887 | Improved |
| smooth maps with 60 arcmin beam instead of 20.6 arcmin | cmbNNCS · CMB-ML test | RMSE: 4.235 -> 0.788 | Improved |
| compare power spectra at high multipole (l>500) | CNILC vs cmbNNCS · CMB-ML power spectrum | power spectrum: qualitative: CNILC overshoot, cmbNNCS undershoot | Mixed |
Closest papers by meaning
- CosmoBench: A Multiscale, Multiview, Multitask Cosmology Benchmark for Geometric Deep Learning (NeurIPS 2025 · similarity 0.85)
- The Multimodal Universe: Enabling Large-Scale Machine Learning with 100 TB of Astronomical Scientific Data (NeurIPS 2024 · similarity 0.83)
- VisRec: A Semi-Supervised Approach to Visibility Data Reconstruction in Radio Astronomy (AAAI 2025 · similarity 0.81)
- MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception (NeurIPS 2026 · similarity 0.80)
- Revealing the 3D Cosmic Web through Gravitationally Constrained Neural Fields (ICLR 2025 · similarity 0.80)
- A systematic evaluation of vision-language models for observational astronomical reasoning tasks (NeurIPS 2026 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
