Nature Machine Intelligence

Article	Citations
Physical benchmarks for testing algorithms	961
A challenge for the law and artificial intelligence	845
Towards reproducible robotics research	626
Author Correction: Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	622
Artificial intelligence tackles the nature–nurture debate	610
Tailored structured peptide design with a key-cutting machine approach	504
Identifying spatial single-cell-level interactions with graph transformer	490
A multi-modal deep language model for contaminant removal from metagenome-assembled genomes	430
From embodied intelligence to physical AI	427
A domain-adapted large language model to support clinicians in psychiatric clinical practice	399
Discussions of machine versus living intelligence need more clarity	368
A soft-packaged and portable rehabilitation glove capable of closed-loop fine motor skills	351
Author Correction: A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions	339
Neural sampling from cognitive maps enables goal-directed imagination and planning	261
Wing-strain-based flight control of flapping-wing drones through reinforcement learning	243
A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit	227
Physically constrained generative adversarial networks for improving precipitation fields from Earth system models	217
Investigating machine moral judgement through the Delphi experiment	213
Materiality and risk in the age of pervasive AI sensors	205
A question of trust for AI research in medicine	199
Deep neural networks with controlled variable selection for the identification of putative causal genetic variants	192
AI pioneers win 2024 Nobel prizes	184
Large language models still struggle with false beliefs	182
Reusability report: Deep learning-based analysis of images and spectroscopy data with AtomAI	180
Advancing ethics review practices in AI research	172

Robust virtual staining of landmark organelles with Cytoland	165
Inferring spatial single-cell-level interactions through interpreting cell state and niche correlations learned by self-supervised graph transformer	163
Fast and generalizable micromagnetic simulation with deep neural nets	162
A Global South perspective for ethical algorithms and the State	158
Transformer-based protein generation with regularized latent space optimization	155
Pseudodata-based molecular structure generator to reveal unknown chemicals	154
Maximum diffusion reinforcement learning	154
Are neural network representations universal or idiosyncratic?	154
Zero-shot transfer of protein sequence likelihood models to thermostability prediction	153
The curious case of the test set AUROC	152
Recurrent graph optimal transport for learning 3D flow motion in particle tracking	151
Reshaping the discovery of self-assembling peptides with generative AI guided by hybrid deep learning	151
Direct conformational sampling from peptide energy landscapes through hypernetwork-conditioned diffusion	149
Quantum circuit optimization with AlphaTensor	147
Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene	145
Generative AI for designing and validating easily synthesizable and structurally novel antibiotics	143
What’s the next word in large language models?	142
How to break information cocoons	142
LLMs displaying less cognitive bias are not necessarily better decision makers	140
A new perspective on the simulation of stochastic problems in fluid mechanics with diffusion models	140
Codon language embeddings provide strong signals for use in protein engineering	137
Learning from models beyond fine-tuning	137
Deep spectral component filtering as a foundation model for spectral analysis demonstrated in metabolic profiling	137
Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	132
Multi-animal 3D social pose estimation, identification and behaviour embedding with a few-shot learning framework	132
Empowering biomedical evidence exploration and synthesis with deep knowledge graph research	131
Guiding generative models to uncover diverse and novel crystals via reinforcement learning	129
Laplace neural operator for solving differential equations	129
Error-controlled non-additive interaction discovery in machine learning models	128
Machine learning prediction of enzyme optimum pH	120
AI reality check	118
The TRIPOD-P reporting guideline for improving the integrity and transparency of predictive analytics in healthcare through study protocols	117
What is in your LLM-based framework?	115
Seeking a quantum advantage for machine learning	114
Bridging peptide presentation and T cell recognition with multi-task learning	114
Collaborative creativity in AI	114
Foundation models in healthcare require rethinking reliability	110
Accurate and robust protein sequence design with CarbonDesign	110
A personalized time-resolved 3D mesh generative model for unveiling normal heart dynamics	109
Deciphering RNA–ligand binding specificity with GerNA-Bind	107
Deep learning supported discovery of biomarkers for clinical prognosis of liver cancer	103
Image-based generation for molecule design with SketchMol	102
Tandem mass spectrum prediction for small molecules using graph transformers	100
Unifying multi-sample network inference from prior knowledge and omics data with CORNETO	99
Unsupervised learning of topological non-Abelian braiding in non-Hermitian bands	98
Life-threatening ventricular arrhythmia detection challenge in implantable cardioverter–defibrillators	98
A multi-modal pre-training transformer for universal transfer learning in metal–organic frameworks	96
Learning high-level visual representations from a child’s perspective without strong inductive biases	96
Uncertainty-guided dual-views for semi-supervised volumetric medical image segmentation	96
LLM-based agentic systems in medicine and healthcare	95

Human–AI adaptive dynamics drives the emergence of information cocoons	94
Morphological flexibility in robotic systems through physical polygon meshing	94
Neural scaling of deep chemical models	94
Multimodal learning with graphs	93
Algorithmic fairness and bias mitigation for clinical machine learning with deep reinforcement learning	92
Efficient generation of protein pockets with PocketGen	92
Advanced AI assistants that act on our behalf may not be ethically or legally feasible	92
Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering	91
Reusability report: Exploring the transferability of self-supervised learning models from single-cell to spatial transcriptomics	91
Foundation models and the privatization of public knowledge	91
Writing the rules in AI-assisted writing	89
Anniversary AI reflections	89
ARNLE model identifies prevalence potential of SARS-CoV-2 variants	88
Differentiable visual computing for inverse problems and machine learning	87
Autonomous navigation of intelligent microrobotic swarms in unknown environments	87
A multimodal cell-free RNA language model for liquid biopsy applications	86
Enhancing deep learning-based field reconstruction with a differentiable learning framework	85
A method for multiple-sequence-alignment-free protein structure prediction using a protein language model	85
ResGen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling	85
Reconstructing growth and dynamic trajectories from single-cell transcriptomics data	85
Synthetic data accelerates the development of generalizable learning-based algorithms for X-ray image analysis	85
Moving towards genome-wide data integration for patient stratification with Integrate Any Omics	84
Human-behaviour-based social locomotion model improves the humanization of social robots	84
Geometric deep learning reveals the spatiotemporal features of microscopic motion	83
A neuro-vector-symbolic architecture for solving Raven’s progressive matrices	82
From attribution maps to human-understandable explanations through Concept Relevance Propagation	81
Model-based reinforcement learning for ultrasound-driven autonomous microrobots	81
A social network for AI	80
Lessons from a challenge on forecasting epileptic seizures from non-cerebral signals	80
Publisher Correction: A neural machine code and programming framework for the reservoir computer	79
Distinguishing two features of accountability for AI technologies	79
Learning intermediate physical states for inverse metasurface design	79
Successful implementation of the EU AI Act requires interdisciplinary efforts	79
Mode switching in organisms for solving explore-versus-exploit problems	78
Sampling-enabled scalable manifold learning unveils the discriminative cluster structure of high-dimensional data	78
Learning integral operators via neural integral equations	75
Mask-prior-guided denoising diffusion improves inverse protein folding	75
Towards shared embodied intelligence in humanoid robots through optimization, development and testing of the human-aware ergoCub robot	75
Defending ChatGPT against jailbreak attack via self-reminders	74
Leveraging language model for advanced multiproperty molecular optimization via prompt engineering	74
A collaborative constrained graph diffusion model for the generation of realistic synthetic molecules	74
Delineating the effective use of self-supervised learning in single-cell genomics	74
Active learning for optimal intervention design in causal models	73
An interaction-derived graph learning framework for scoring protein–peptide complexes	73
Closed-form continuous-time neural networks	72
Advances, challenges and opportunities in creating data for trustworthy AI	71
Benchmarking AI-powered docking methods from the perspective of virtual screening	71
Artificial intelligence-powered electronic skin	70
CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling	70
On board with COMET to improve omics prediction models	69
Listening in to perceived speech with contrastive learning	67
A new eye on inherited retinal disease	67
The incentive gap in data work in the era of large models	66
Versatile cardiovascular signal generation with a unified diffusion transformer	66
Deep transfer operator learning for partial differential equations under conditional shift	66
Teaching machines to blend electrolyte cocktails	66
Data-driven discovery of movement-linked heterogeneity in neurodegenerative diseases	66
Unsupervised ensemble-based phenotyping enhances discoverability of genes related to left-ventricular morphology	65
Lossless data compression by large models	65
Mitigating the missing-fragmentation problem in de novo peptide sequencing with a two-stage graph-based deep learning model	65
A computational framework for neural network-based variational Monte Carlo with Forward Laplacian	65
Solving sparse finite element problems on neuromorphic hardware	65
Incorporating physics into data-driven computer vision	63
A generalizable deep learning framework for inferring fine-scale germline mutation rate maps	63
Interpretable meta-score for model performance	63
Machine learning-enabled globally guaranteed evolutionary computation	62
A unified deep framework for peptide–major histocompatibility complex–T cell receptor binding prediction	62
Current-diffusion model for metasurface structure discoveries with spatial-frequency dynamics	61
Invalid SMILES are beneficial rather than detrimental to chemical language models	61
Deep learning for predicting rate-induced tipping	60
Large language models challenge the future of higher education	59
Three types of incremental learning	59
Fast, scale-adaptive and uncertainty-aware downscaling of Earth system model fields with generative machine learning	58
Augmenting large language models with chemistry tools	57
Predicting the conformational flexibility of antibody and T cell receptor complementarity-determining regions	57
Publisher Correction: Advancing ethics review practices in AI research	56
Space missions out of this world with AI	55
Synergy-based robotic quadruped leveraging passivity for natural intelligence and behavioural diversity	54
Deconstructing the generalization gap	54
Competing Biases underlie Overconfidence and Underconfidence in LLMs	54

A process-centric manipulation taxonomy for the organization, classification and synthesis of tactile robot skills	53
Automated construction of cognitive maps with visual predictive coding	53
A family of large language models for materials research with insights into model adaptability in continued pretraining	53
Multiscale topology-enabled structure-to-sequence transformer for protein–ligand interaction predictions	52
Predicting the prevalence of complex genetic diseases from individual genotype profiles using capsule networks	52
Discovering neural policies to drive behaviour by integrating deep reinforcement learning agents with biological neural networks	51
Labelling instructions matter in biomedical image analysis	50
The importance of negative training data for robust antibody binding prediction	50
Generalized biological foundation model with unified nucleic acid and protein language	50
Realistic morphology-preserving generative modelling of the brain	49
Visual speech recognition for multiple languages in the wild	48
Aligning generalization between humans and machines	48
Design of prime-editing guide RNAs with deep transfer learning	48
Pan-Peptide Meta Learning for T-cell receptor–antigen binding recognition	48
When large language models are reliable for judging empathic communication	47
Deep learning-based prediction of the selection factors for quantifying selection in immune receptor repertoires	47
Realigning AI technology towards the Sustainable Development Goals	47
Towards a universal model for spin–orbit physics	46
Author Correction: Predicting equilibrium distributions for molecular systems with deep learning	46
Geometric deep learning of particle motion by MAGIK	46
An integrated framework to accelerate protein design through mutagenesis	46
A disease-specific language model for variant pathogenicity in cardiac and regulatory genomics	45
Accelerating protein engineering with fitness landscape modelling and reinforcement learning	45
Reply to: Inability of a graph neural network heuristic to outperform greedy algorithms in solving combinatorial optimization problems	44
Towards unveiling sensitive and decisive patterns in explainable AI with a case study in geometric deep learning	44
Why design choices matter in recommender systems	44
Type II mechanoreceptors and cuneate spiking neuronal network enable touch localization on a large-area e-skin	44
Weak signal extraction enabled by deep neural network denoising of diffraction data	43
A large-scale randomized study of large language model feedback in peer review	43
Parameter-efficient fine-tuning of large-scale pre-trained language models	43
On the caveats of AI autophagy	43
Efficient rare event sampling with unsupervised normalizing flows	42
Embodied large language models enable robots to complete complex tasks in unpredictable environments	42
A ‘programming’ framework for recurrent neural networks	42
Clinical large language models with misplaced focus	42