Research articles

A physics-based training pipeline is developed to help tackle the challenges of data scarcity. The framework aligns large language models to a physically consistent initial state that is fine-tuned for learning polymer properties.

MultiSTAAR provides a general and flexible statistical framework for functionally informed multi-trait rare variant analysis of biobank-scale sequencing studies by jointly analyzing multiple traits and incorporating annotation information.

A multimodal computational framework is proposed to integrate single-cell RNA sequencing data with phenotypic information to map complex genotype–phenotype relationships. This approach helps to refine cellular heterogeneity analysis, identify cross-tissue biomarkers and reveal polyfunctional characteristics of genes with cellular resolution.

This study presents a neuromorphic computing platform capable of learning cross-modal, event-driven signals for efficient real-time knowledge generalization. It also achieves zero-shot transfer learning for multimodal data.

This study shows a viable pathway to the efficient deployment of state-of-the-art large language models using mixture of experts on 3D analog in-memory computing hardware.

A large-scale electronic circular dichroism spectrum dataset is proposed and the ECDFormer framework is developed to achieve accurate and interpretable ECD spectrum prediction for natural products.

This study introduces an extensible framework—Morpho—for shape optimization, enabling researchers to predict the structure of soft materials, such as complex fluids, gels, particulate and biological materials.

A multi-task deep learning method for molecular electronic structures, called MEHnet, is developed to predict various molecular properties in a unified framework, approaching chemical accuracy while exhibiting local DFT-level computational costs.

This study introduces an in-memory deep Bayesian active learning framework that uses the stochastic properties of memristors for in situ probabilistic computations. This framework can greatly improve the efficiency and speed of artificial intelligence learning tasks, as demonstrated with a robot skill-learning task.

A deep learning framework is proposed with real-world pharmacovigilance data to predict population-scale toxicity profiles of checkpoint inhibitor immunotherapy, enabling proactive toxicity monitoring and timely tailoring of treatment.

This work presents a graph signal processing method, gene signal pattern analysis, to embed gene signals from single-cell sequencing data. In diverse experimental set-ups and case studies, GSPA establishes a gene-based framework for single-cell analysis.

The spatiotemporal style transfer (STST) algorithm enables video generation by selectively manipulating the spatial and temporal features of natural videos, fostering vision science research in both biological and artificial systems.

Spatial Modeling Algorithms for Reactions and Transport (SMART) is a software package that allows users to simulate spatially resolved biochemical signaling networks within realistic geometries of cells and organelles.

The Digital Brain platform is capable of simulating spiking neuronal networks at the neuronal scale of the human brain. The platform is used to reproduce blood-oxygen-level-dependent signals in both the resting state and action, thereby predicting the visual evaluation scores.

This study introduces a²c, a computational method that leverages machine learning and atomistic simulations to predict the most likely crystallization products upon annealing of amorphous precursors. The a²c tool was demonstrated on a variety of materials, including oxides, nitrides and metallic glasses, and can assist researchers in discovering synthesis pathways for materials design.

BAAIWorm is an integrative data-driven model of C. elegans that simulates interactions between the brain, body and environment. The biophysically detailed neuronal model is capable of replicating the zigzag movement observed in this species.

Researchers show that large language models exhibit social identity biases similar to humans, having favoritism toward ingroups and hostility toward outgroups. These biases persist across models, training data and real-world human–LLM conversations.

This study proposes an algorithm for generating randomized networks that preserve the weighted degree sequence. The procedure outperforms standard rewiring algorithms and extends to multiple network types, including directed and signed networks.

This work presents an artificial intelligence framework to learn geometry-dependent solution operators of partial differential equations (PDEs). The framework enables scalable and fast approximations of PDE solutions on a variety of 3D geometries.

This work applies diffusion models to conditional molecule generation and shows how they can be used to tackle various structure-based drug design problems