# CrystalJev: calibrated foundation-model probabilities decide what to compute next in materials discovery

arXiv: https://arxiv.org/abs/2610.06985 · HTML: https://mlsanity.com/paper/2610.06985/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 53 references that are in the map.
- First posted on arXiv 4 Oct 2026.

## Builds on (most cited first, 8 of 8)

- [UMA: A Family of Universal Models for Atoms](https://mlsanity.com/paper/2506.23971/) (2025)
- [Thinking Fast and Slow in AI](https://mlsanity.com/paper/2010.06002/) (2020)
- [Selection by Prediction with Conformal p-values](https://mlsanity.com/paper/2210.01408/) (2022)
- [Materials Property Prediction with Uncertainty Quantification: A Benchmark Study](https://mlsanity.com/paper/2211.02235/) (2022)
- [MatterTune: An Integrated, User-Friendly Platform for Fine-Tuning Atomistic Foundation Models to Accelerate Materials Simulation and Discovery](https://mlsanity.com/paper/2504.10655/) (2025)
- [Crys-JEPA: Accelerating Crystal Discovery via Embedding Screening and Generative Refinement](https://mlsanity.com/paper/2605.14759/) (2026)
- [Knowing when to trust machine-learned interatomic potentials](https://mlsanity.com/paper/2605.00640/) (2026)
- [When Active Learning Fails, Uncalibrated Out of Distribution Uncertainty Quantification Might Be the Problem](https://mlsanity.com/paper/2511.17760/) (2025)

## Abstract (arXiv)

Atomistic foundation models screen hypothetical crystals by thresholding predicted energies, an output that states neither how likely a candidate is to be stable nor whether more computation would change the decision. CrystalJev reads one forward pass of a frozen potential on the unrelaxed structure and returns calibrated probabilities, conformal guarantees and a rule for escalating candidates to relaxation or density-functional theory (DFT). Across 65 Matbench Discovery models, threshold calls hide cost assumptions, and ordering slower computation by its value for the decision, rather than by probability, recovers most of its benefit at a tenth of its cost. In a registered prospective test with 2,100 DFT calculations on ionic substitutions, forecasts calibrated on existing data over-stated the stable fraction of new candidates by 0.9--2.2 percentage points, because the population had shifted, not the labelling protocol; a few hundred labels from the new pool removed this bias more reliably than label-free corrections.