# Li-P-S Electrolyte Materials as a Benchmark for Machine-Learned Interatomic Potentials

Preprint · Natascia L. Fragapane, Volker L. Deringer

- Archive id: arxiv:2511.16569
- Published: 2025-11-20
- Venue: PubMed Central
- arXiv: https://arxiv.org/abs/2511.16569
- Citations: 0
- Source URL: https://arxiv.org/abs/2511.16569
- PDF: https://arxiv.org/pdf/2511.16569
- Provenance: arxiv

## Abstract

With the growing availability of machine-learned interatomic potential (MLIP) models for materials simulations, there is an increasing demand for robust, automated, and chemically insightful benchmarking methodologies. In response, we here introduce LiPS-25, a curated benchmark dataset for a canonical series of solid-state electrolyte materials from the Li2S-P2S5 pseudo-binary compositional line, including crystalline and amorphous configurations. Together with the dataset, we present a suite of performance tests that range from conventional numerical error metrics to physically motivated evaluation tasks. With a focus on graph-based MLIP architectures, we run numerical experiments that assess (i) the effect of hyperparameters and (ii) the fine-tuning behavior of selected pre-trained ("foundational") MLIP models. Beyond the Li-P-S solid-state electrolytes, we expect that such benchmarks and their code implementations can be readily adapted to other material systems.

_Abstract reproduced under CC0-1.0 (arXiv descriptive metadata)._

Topics: [AI for Science](https://space-kitty.com/topics/ai-for-science)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.