USRA details NASA-IBM lunar foundation model fusing 11 lunar data modalities
victormustar · x · 2026-09-19
A USRA press release adds context on the NASA-IBM Lunar Foundation Model (Sept 18, 2026).
- Developed in a collaboration led by NASA and IBM Research, with USRA scientist Rachel Slank contributing planetary science expertise, dataset development, and scientific evaluation
- SomBench fuses imagery, topography, illumination geometry, thermophysical properties, mineralogy, radar, gravity and other lunar data into 2 million co-registered bundles
- In a demo task, the model's lunar polar ice prospectivity maps preserve fine-scale patterns of reference data, outperforming a ConvNeXt baseline
- The goal is helping researchers analyze large, diverse datasets from upcoming lunar missions
Related event: NASA and IBM Open-Source Lunar Foundation Model(3 posts)→
More from Research
- Nearly 20 openjev models cataloged as hobbyist preps first community leaderboard — airesearch12 · 2026-09-19
- Mnemos engine builds synaptic-like agent memory via resonant engrams — RileyRalmuto · 2026-09-19
- Shared paper claims AI models have a concept of pain and avoid harm — skolnaja · 2026-09-19
- Steering Vectors as a Softer Way to Limit Reasoning Budgets? — maddie-lovelace · 2026-09-19
- MMLU-Redux paper finds ~6.5% of MMLU questions contain labeling errors — PMinervini · 2026-09-19
- NASA and IBM open-source lunar foundation model pretrained on 2M multimodal tiles — victormustar · 2026-09-19