About

Bingqing Cheng, born in 1991, is an Assistant Professor of Chemistry at the University of California, Berkeley, starting January 2024. He holds a B.Eng. in Mechanical Engineering from The University of Hong Kong and Shanghai Jiao Tong University (2012), an M. Phil. in Materials Science from The University of Hong Kong (2014), and a Ph.D. in Materials Science from EPFL (2019). Cheng has held positions as a Junior Research Fellow at Trinity College, University of Cambridge, and as a Departmental Early Career Fellow at the University of Cambridge. He was an Assistant Professor at the Institute of Science and Technology Austria before joining UC Berkeley. His research focuses on theoretical and physical chemistry, materials, polymers, and nanoscience. Cheng's group develops methods to extend the scope of atomistic simulations to understand and predict materials properties that are difficult to access with traditional approaches. His work involves deploying and designing techniques such as machine learning, enhanced sampling, path-integral molecular dynamics, and free energy estimation. The systems studied include energy materials, aqueous systems, and matter under extreme conditions, with the goal of advancing society through education and research in the chemical sciences.

Research topics

• Computer Science
• Machine Learning
• Data Mining
• Artificial Intelligence
• Physics
• Engineering
• Astrophysics
• Management science
• Nanotechnology
• Biochemical engineering
• Data science
• Chemistry
• Thermodynamics
• Materials science
• Chemical physics

Selected publications

• Long-range electrostatics for machine learning interatomic potentials is easier than we thought

  The Journal of Chemical Physics · 2026-02-12 · 1 citations

  articleOpen accessSenior author

  The lack of long-range electrostatics is a key limitation of modern machine learning interatomic potentials (MLIPs), hindering reliable applications to interfaces, charge-transfer reactions, polar and ionic materials, and biomolecules. In this Perspective, we distill two design principles behind the Latent Ewald Summation framework, which can capture long-range interactions, charges, and electrical response just by learning from standard energy and force training data: (i) use a Coulomb functional form with environment-dependent charges to capture electrostatic interactions, and (ii) avoid explicit training on ambiguous density functional theory partial charges. When both principles are satisfied, substantial flexibility remains: essentially any short-range MLIP can be augmented; charge equilibration schemes can be added when desired; dipoles and Born effective charges can be inferred or fine-tuned; and charge/spin-state embeddings or tensorial targets can be further incorporated. We also discuss current limitations and open challenges. Together, these minimal, physics-guided design rules suggest that incorporating long-range electrostatics into MLIPs is simpler and perhaps more broadly applicable than is commonly assumed.

  PublisherOA PDFDOI

• Systematic Trends in Water Properties Across Jacob's Ladder Density Functionals

  ChemRxiv · 2026-03-05

  articleOpen accessSenior author

  Kohn-Sham density functional theory is the workhorse of computational materials science and chemistry, but the optimal choice of exchange-correlation approximation for a given system is often unclear. Traditional benchmark studies typically compare selected density functional approximations (DFAs) against high-level quantum chemistry results for small structures and static energies, rather than against experimental thermodynamic or response properties. Here, we show how machine learning interatomic potentials (MLIPs) allow for molecular dynamics simulations of these properties at scale, thereby providing a thermodynamic benchmark for DFAs. Focusing on the wellstudied but challenging case of bulk water at ambient conditions, we benchmark 11 DFAs spanning the top four rungs of Jacob's ladder (GGA, meta-GGA, hybrid, and double hybrid) with and without empirical dispersion corrections. Using MLIPs trained on each DFA, we compute infrared spectra, heat capacities, radial distribution functions, and diffusivities. We find a general improvement up Jacob's ladder towards double hybrids, though there is substantial variation between DFAs within the same rung and across different properties. Dispersion corrections yield only a modest influence on the computed properties. We elucidate the similarities and hierarchy among DFAs, both based on correlating the DFA energies and forces for the same training configurations, and based on analyzing the finite-temperature observables. Our work thus demonstrates the power of MLIPs for finite-temperature benchmarking of electronic structure methods.

  PublisherOA PDFDOI

• Scalable Multitemperature Free Energy Sampling of Classical Ising Spin States

  ArXiv.org · 2025-03-11

  preprintOpen accessSenior author

  Generative models have advanced significantly in sampling material systems with continuous variables, such as atomistic structures. However, their application to discrete variables, like atom types or spin states, remains underexplored. In this work, we introduce a discrete flow matching model, tailored for systems with discrete phase-space coordinates (e.g., the Ising model or a multicomponent system on a lattice). This approach enables a single model to sample free energy surfaces over a wide temperature range with minimal training overhead, and the model generation is scalable to larger lattice sizes than those in the training set. We demonstrate our approach on the 2D Ising model, showing efficient and reliable free energy sampling. These results highlight the potential of flow matching for low-cost, scalable free energy sampling in discrete systems and suggest promising extensions to alchemical degrees of freedom in crystalline materials. The codebase developed for this work is openly available at https://github.com/tuoping/alchemicalFES.

  PublisherOA PDFDOI

• Machine learning interatomic potential can infer electrical response

  ArXiv.org · 2025-04-07 · 2 citations

  preprintOpen accessSenior author

  Modeling the response of material and chemical systems to electric fields remains a longstanding challenge. Machine learning interatomic potentials (MLIPs) offer an efficient and scalable alternative to quantum mechanical methods but do not by themselves incorporate electrical response. Here, we show that polarization and Born effective charge (BEC) tensors can be directly extracted from long-range MLIPs within the Latent Ewald Summation (LES) framework, solely by learning from energy and force data. Using this approach, we predict the infrared spectra of bulk water under zero or finite external electric fields, ionic conductivities of high-pressure superionic ice, and the phase transition and hysteresis in ferroelectric PbTiO$_3$ perovskite. This work thus extends the capability of MLIPs to predict electrical response--without training on charges or polarization or BECs--and enables accurate modeling of electric-field-driven processes in diverse systems at scale.

  PublisherOA PDFDOI

• Latent Ewald summation for machine learning of long-range interactions

  npj Computational Materials · 2025-03-26 · 61 citations

  articleOpen access1st authorCorresponding

  Abstract Machine learning interatomic potentials (MLIPs) often neglect long-range interactions, such as electrostatic and dispersion forces. In this work, we introduce a straightforward and efficient method to account for long-range interactions by learning a hidden variable from local atomic descriptors and applying an Ewald summation to this variable. We demonstrate that in systems including charged and polar molecular dimers, bulk water, and water-vapor interface, standard short-ranged MLIPs can lead to unphysical predictions even when employing message passing. The long-range models effectively eliminate these artifacts, with only about twice the computational cost of short-range MLIPs.

  PublisherOA PDFDOI

• A Universal Augmentation Framework for Long-Range Electrostatics in Machine Learning Interatomic Potentials

  Journal of Chemical Theory and Computation · 2025-12-10 · 10 citations

  articleOpen accessSenior authorCorresponding

  Most current machine learning interatomic potentials (MLIPs) rely on short-range approximations, without explicit treatment of long-range electrostatics. To address this, we recently developed the Latent Ewald Summation (LES) method, which infers electrostatic interactions, polarization, and Born effective charges (BECs), just by learning from energy and force training data. Here, we present LES as a standalone library, compatible with any short-range MLIP, and demonstrate its integration with methods such as MACE, NequIP, Allegro, CACE, CHGNet, and UMA. We benchmark LES-enhanced models on distinct systems, including bulk water, polar dipeptides, and gold dimer adsorption on defective substrates, and show that LES not only captures correct electrostatics but also improves accuracy. Additionally, we scale LES to large and chemically diverse data by training MACELES-OFF on the SPICE set containing molecules and clusters, making a universal MLIP with electrostatics for organic systems, including biomolecules. MACELES-OFF is more accurate than its short-range counterpart (MACE-OFF) trained on the same data set, predicts dipoles and BECs reliably, and has better descriptions of bulk liquids. By enabling efficient long-range electrostatics without directly training on electrical properties, LES paves the way for electrostatic foundation MLIPs.

  PublisherOA PDFDOI

• Ion-modulated structure, proton transfer, and capacitance in the Pt(111)/water electric double layer

  ArXiv.org · 2025-09-17 · 1 citations

  preprintOpen accessSenior author

  The electric double layer (EDL) governs electrocatalysis, energy conversion, and storage, yet its atomic structure, capacitance, and reactivity remain elusive. Here we introduce a machine learning interatomic potential framework that incorporates long-range electrostatics, enabling nanosecond simulations of metal-electrolyte interfaces under applied electric bias with near-quantum-mechanical accuracy. At the benchmark Pt(111)/water and Pt(111)/aqueous KF electrolyte interfaces, we resolve the molecular structure of the EDL, reveal proton-transfer mechanisms underlying anodic water dissociation and the diffusion of ionic water species, and compute differential capacitance. We find that the nominally inert K+ and F- ions, while leaving interfacial water structure largely unchanged, screen bulk fields, slow proton transfer, and generate a prominent capacitance peak near the potential of zero charge. These results show that ion-specific interactions, which are ignored in mean-field models, are central to capacitance and reactivity, providing a molecular basis for interpreting experiments and designing electrolytes.

  PublisherOA PDFDOI

• Thermal transport of amorphous hafnia across the glass transition

  ArXiv.org · 2025-02-05 · 1 citations

  preprintOpen accessSenior author

  Heat transport in glasses across a wide range of temperature is vital for applications in gate dielectrics and heat insulator. However, it remains poorly understood due to the challenges of modeling vibrational anharmonicity below glass transition temperature and capturing configurational dynamics across the transition. Interestingly, recent calculations predicted that amorphous hafnia (a-HfO$_2$) exhibits an unusual drop in thermal conductivity ($κ$) with temperature, contrasting with the typical rise or saturation observed in glasses upon heating. Using molecular dynamics simulations with a machine-learning-based neuroevolution potential, we compute the vibrational properties and $κ$ of a-HfO$_2$ from 50~K to 2000~K. At low temperatures, we employ the Wigner transport equation to incorporate both anharmonicity and Bose-Einstein statistics of atomic vibration in the calculation of $κ$. At above 1200~K, atomic diffusion breaks down the Lorentzian-shaped quasiparticle picture and makes the lattice-dynamics treatment invalid. We thus use molecular dynamics with the Green-Kubo method to capture convective heat transport in a-HfO$_2$ near the glass transition at around 1500~K. Additionally, by extending the Wigner transport equation to supercooled liquid states, we find the crucial role of low-frequency modes in facilitating heat convection. The computed $κ$ of a-HfO$_2$, based on both Green-Kubo and Wigner transport theories, reveals a continuous increase with temperature up to 2000~K.

  PublisherOA PDFDOI

• Lattice distortion leads to glassy thermal transport in crystalline Cs ₃ Bi ₂ I ₆ Cl ₃

  Proceedings of the National Academy of Sciences · 2025-10-06 · 3 citations

  articleOpen accessSenior authorCorresponding

  The glassy thermal conductivities observed in crystalline inorganic perovskites such as Cs 3 Bi 2 I 6 Cl 3 are perplexing and lacking theoretical explanations. Here, we first experimentally measure its thermal transport behavior from 20 to 300 K, after synthesizing Cs 3 Bi 2 I 6 Cl 3 single crystals. Using path-integral molecular dynamics simulations driven by machine learning potentials, we reveal that Cs 3 Bi 2 I 6 Cl 3 has large lattice distortions at low temperatures, which may be related to the large atomic size mismatch. Employing the Wigner formulation of thermal transport, we reproduce the experimental thermal conductivities based on lattice-distorted structures. This study thus provides a framework for predicting and understanding glassy thermal transport in materials with strong lattice disorder.

  PublisherOA PDFDOI

• Automatic feature selection and weighting in molecular systems using Differentiable Information Imbalance

  Nature Communications · 2025-01-02 · 10 citations

  articleOpen access

  Feature selection is essential in the analysis of molecular systems and many other fields, but several uncertainties remain: What is the optimal number of features for a simplified, interpretable model that retains essential information? How should features with different units be aligned, and how should their relative importance be weighted? Here, we introduce the Differentiable Information Imbalance (DII), an automated method to rank information content between sets of features. Using distances in a ground truth feature space, DII identifies a low-dimensional subset of features that best preserves these relationships. Each feature is scaled by a weight, which is optimized by minimizing the DII through gradient descent. This allows simultaneously performing unit alignment and relative importance scaling, while preserving interpretability. DII can also produce sparse solutions and determine the optimal size of the reduced feature space. We demonstrate the usefulness of this approach on two benchmark molecular problems: (1) identifying collective variables that describe conformations of a biomolecule, and (2) selecting features for training a machine-learning force field. These results show the potential of DII in addressing feature selection challenges and optimizing dimensionality in various applications. The method is available in the Python library DADApy.

  PublisherOA PDFDOI

Frequent coauthors

• Michele Ceriotti

  École Polytechnique Fédérale de Lausanne

  21 shared

• Chris J. Pickard

  15 shared

• Mandy Bethkenhagen

  Université Paris 8

  14 shared

• Tiesheng Dou

  12 shared

• Xiaonong Dong

  China Institute of Water Resources and Hydropower Research

  12 shared

• A.H.W. Ngan

  8 shared

• Aleks Reinhardt

  University of Cambridge

  7 shared

• Gareth A. Tribello

  7 shared

Education

• Ph.D.

  École Polytechnique Fédérale de Lausanne

  2019