Lagrangian Neural Networks (LNNs) use neural networks to parameterize arbitrary Lagrangians, enabling energy-conserving learned dynamics without canonical coordinates. Unlike Hamiltonian approaches, LNNs handle relativistic systems and extend to graphs via Lagrangian Graph Networks.
Frey et al. discover empirical power-law scaling relations for both chemical language models (ChemGPT, up to 1B parameters) and equivariant GNN interatomic potentials, finding that neither domain has saturated with respect to model size, data, or compute.
Tartarus introduces a modular suite of realistic molecular design benchmarks grounded in computational chemistry simulations. Benchmarking eight generative models reveals that no single algorithm dominates all tasks, and simple genetic algorithms often outperform deep generative models.
This paper replaces discrete network layers with continuous ordinary differential equations (ODEs), allowing for adaptive computation depth and constant memory cost during training via the adjoint sensitivity method. It introduces Continuous Normalizing Flows and latent ODEs for time-series.
This paper introduces DynamicFlow, a full-atom stochastic flow matching model that simultaneously generates ligand molecules and transforms protein pockets from apo to holo states. It also contributes a new dataset of MD-simulated apo-holo pairs derived from MISATO.
MOFFlow is the first deep generative model tailored for Metal-Organic Framework (MOF) structure prediction. It utilizes Riemannian flow matching on SE(3) to assemble rigid building blocks (metal nodes and organic linkers), achieving higher accuracy and scalability than atom-based methods on large systems.
Sussman and Wisdom’s 1992 study used the Supercomputer Toolkit and symplectic mapping to integrate the entire Solar System for 100 million years, confirming chaotic behavior with an exponential divergence timescale of ~4 million years and demonstrating that long-term planetary motion is fundamentally unpredictable.
This paper bridges Molecular Dynamics and Transition State Theory by applying a dynamical corrections formalism to surface diffusion, identifying a low-temperature bounce-back mechanism causing non-Arrhenius behavior.
This 1994 handbook chapter serves as a practical user guide for the Embedded-Atom Method (EAM). It details the theoretical derivation from density-functional theory, synthesizes related methods like the Glue Model, and provides a complete tutorial on fitting potentials, illustrated with a specific implementation for the Ni-Al-B system.
This 1993 review systematizes the Embedded-Atom Method (EAM) as a practical semi-empirical approach for metallic systems. It synthesizes theory, applications, and connections to related methods while addressing the limitations of pair potentials.
This paper validates the homogeneous Evans method for calculating thermal conductivity against experimental Argon data. It demonstrates broad agreement across the phase diagram but identifies significant non-monotonic behavior and enhanced long-time tails near the critical point.
This paper resolves Levinthal’s paradox by replacing the single-pathway view with a statistical energy landscape approach. It introduces the concepts of the folding funnel, the glass transition in proteins, and the ‘stability gap’ as a design principle for foldable sequences.