arXiv subjects

Haote Li

Publications and source records attributed to Haote Li.

4 recordsLinked to original sources

arXiv2026

RxnCLF: Contrastive Transformation-Aware Reaction Foundation Model for Improved Reactivity Prediction

Reaction yield prediction remains challenging because labeled data are scarce and reaction space is both combinatorially large and sparsely populated, limiting the generalization of existing reaction representations. String-, fingerprint-, and graph-based reaction encodings only partially capture chemical transformations, making accurate prediction difficult for reactions with complex substrates. We propose reaction contrastive learning foundation (RxnCLF), a self-supervised contrastive framework for reaction representation learning. RxnCLF is built on a condensed reaction graph (CRG) that unifies reactant and product information into a single graph, enabling the model to learn explicit and enriched transformation structure rather than disconnected graphs. Pretrained on 1.7 million Pistachio reactions, RxnCLF learns a compact and continuous latent space that captures both reaction-center features and broader side chain contexts, making it transformation-aware and chemically interpretable. Fine-tuned on multiple yield prediction benchmarks, including Buchwald-Hartwig, Pd-catalyzed BH coupling, and proprietary HTE C-N coupling and amide formation datasets, RxnCLF consistently outperforms graph and sequence-based baselines, improving R2 and achieving the best performance overall. Our results highlight the promise of CRG-based RxnCLF as a scalable reaction foundation model, with the potential to generalize across broader reaction spaces and support diverse downstream reaction informatics tasks, including regioselectivity prediction, enantioselectivity prediction, and reaction condition optimization.

cs.LG↗

arXiv2024

DirectMultiStep: Direct Route Generation for Multistep Retrosynthesis

Traditional computer-aided synthesis planning (CASP) methods rely on iterative single-step predictions, leading to exponential search space growth that limits efficiency and scalability. We introduce a series of transformer-based models, that leverage a mixture of experts approach to directly generate multistep synthetic routes as a single string, conditionally predicting each transformation based on all preceding ones. Our DMS Explorer XL model, which requires only target compounds as input, outperforms state-of-the-art methods on the PaRoutes dataset with 1.9x and 3.1x improvements in Top-1 accuracy on the n$_1$ and n$_5$ test sets, respectively. Providing additional information, such as the desired number of steps and starting materials, enables both a reduction in model size and an increase in accuracy, highlighting the benefits of incorporating more constraints into the prediction process. The top-performing DMS-Flex (Duo) model scores 25-50% higher on Top-1 and Top-10 accuracies for both n$_1$ and n$_5$ sets. Additionally, our models successfully predict routes for FDA-approved drugs not included in the training data, demonstrating strong generalization capabilities. While the limited diversity of the training set may affect performance on less common reaction types, our multistep-first approach presents a promising direction towards fully automated retrosynthetic planning.

cs.LG↗

arXiv2024

Simulating electronic structure on bosonic quantum computers

Quantum harmonic oscillators, or qumodes, provide a promising and versatile framework for quantum computing. Unlike qubits, which are limited to two discrete levels, qumodes have an infinite-dimensional Hilbert space, making them well-suited for a wide range of quantum simulations. In this work, we focus on the molecular electronic structure problem. We propose an approach to map the electronic Hamiltonian into a qumode bosonic problem that can be solved on bosonic quantum devices using the variational quantum eigensolver (VQE). Our approach is demonstrated through the computation of ground potential energy surfaces for benchmark model systems, including H$_2$ and the linear H$_4$ molecule. The preparation of trial qumode states and the computation of expectation values leverage universal ansatzes based on the echoed conditional displacement (ECD), or the selective number-dependent arbitrary phase (SNAP) operations. These techniques are compatible with circuit quantum electrodynamics (cQED) platforms, where microwave resonators coupled to superconducting transmon qubits can offer an efficient hardware realization. This work establishes a new pathway for simulating many-fermion systems, highlighting the potential of hybrid qubit-qumode quantum devices in advancing quantum computational chemistry.

quant-ph↗

arXiv2023

Kernel-Elastic Autoencoder for Molecular Design

We introduce the Kernel-Elastic Autoencoder (KAE), a self-supervised generative model based on the transformer architecture with enhanced performance for molecular design. KAE is formulated based on two novel loss functions: modified maximum mean discrepancy and weighted reconstruction. KAE addresses the long-standing challenge of achieving valid generation and accurate reconstruction at the same time. KAE achieves remarkable diversity in molecule generation while maintaining near-perfect reconstructions on the independent testing dataset, surpassing previous molecule-generating models. KAE enables conditional generation and allows for decoding based on beam search resulting in state-of-the-art performance in constrained optimizations. Furthermore, KAE can generate molecules conditional to favorable binding affinities in docking applications as confirmed by AutoDock Vina and Glide scores, outperforming all existing candidates from the training dataset. Beyond molecular design, we anticipate KAE could be applied to solve problems by generation in a wide range of applications.

cs.LG↗