SPICE

Dataset Card for SPICED Dataset Summary The Scientific Paraphrase and Information ChangE Dataset (SPICED) is a dataset of paired scientific findings from scientific papers, news media, and Twitter. The types of pairs are between <paper, news> and <paper, tweet>. Each pair is labeled for the degree of information simila

Papers using SPICE (9)

A universal augmentation framework for long-range electrostatics in machine learning interatomic potentials2025 · 24 citesHow Accurate Are DFT Forces? Unexpectedly Large Uncertainties in Molecular Datasets2025 · 1 citesCutting Through the Noise: On-the-fly Outlier Detection for Robust Training of Machine Learning Interatomic Potentials2026Molecular electrostatic potentials from machine learning models for dipole and quadrupole predictions2026Power law attention biases for molecular transformers2025A Scalable and Quantum-Accurate Foundation Model for Biomolecular Force Field via Linearly Tensorized Quadrangle Attention2025Nutmeg and SPICE: Models and Data for Biomolecular Machine Learning2024 · 35 citesFrom Molecules to Materials: Pre-training Large Generalizable Models for Atomic Property Prediction2023 · 22 citesSPICE, A Dataset of Drug-like Molecules and Peptides for Training Machine Learning Potentials2022 · 13 cites