Nature Machine Intelligence

Article	Citations
Physical benchmarks for testing algorithms	881
A challenge for the law and artificial intelligence	750
Towards reproducible robotics research	568
Author Correction: A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions	557
Author Correction: Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	533
Artificial intelligence tackles the nature–nurture debate	471
A multi-modal deep language model for contaminant removal from metagenome-assembled genomes	434
From embodied intelligence to physical AI	396
Tailored structured peptide design with a key-cutting machine approach	391
Identifying spatial single-cell-level interactions with graph transformer	386
Discussions of machine versus living intelligence need more clarity	340
A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit	313
Materiality and risk in the age of pervasive AI sensors	306
Physically constrained generative adversarial networks for improving precipitation fields from Earth system models	241
A domain-adapted large language model to support clinicians in psychiatric clinical practice	220
Investigating machine moral judgement through the Delphi experiment	207
A soft-packaged and portable rehabilitation glove capable of closed-loop fine motor skills	198
Wing-strain-based flight control of flapping-wing drones through reinforcement learning	195
A question of trust for AI research in medicine	188
Zero-shot transfer of protein sequence likelihood models to thermostability prediction	181
Recurrent graph optimal transport for learning 3D flow motion in particle tracking	180
Direct conformational sampling from peptide energy landscapes through hypernetwork-conditioned diffusion	167
Deep neural networks with controlled variable selection for the identification of putative causal genetic variants	163
Are neural network representations universal or idiosyncratic?	156
AI pioneers win 2024 Nobel prizes	154

Large language models still struggle with false beliefs	152
Fast and generalizable micromagnetic simulation with deep neural nets	151
Transformer-based protein generation with regularized latent space optimization	148
Robust virtual staining of landmark organelles with Cytoland	145
Pseudodata-based molecular structure generator to reveal unknown chemicals	145
Reusability report: Deep learning-based analysis of images and spectroscopy data with AtomAI	141
A Global South perspective for ethical algorithms and the State	140
Quantum circuit optimization with AlphaTensor	140
Advancing ethics review practices in AI research	140
Reshaping the discovery of self-assembling peptides with generative AI guided by hybrid deep learning	139
Maximum diffusion reinforcement learning	138
Bringing artificial intelligence to business management	138
Generative AI for designing and validating easily synthesizable and structurally novel antibiotics	137
The curious case of the test set AUROC	133
Inferring spatial single-cell-level interactions through interpreting cell state and niche correlations learned by self-supervised graph transformer	132
A new perspective on the simulation of stochastic problems in fluid mechanics with diffusion models	131
Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene	131
What’s the next word in large language models?	131
Error-controlled non-additive interaction discovery in machine learning models	131
Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	123
LLMs displaying less cognitive bias are not necessarily better decision makers	122
How to break information cocoons	121
Learning from models beyond fine-tuning	120
Deep spectral component filtering as a foundation model for spectral analysis demonstrated in metabolic profiling	120
Machine learning prediction of enzyme optimum pH	119
Laplace neural operator for solving differential equations	116
Codon language embeddings provide strong signals for use in protein engineering	114
AI reality check	113
Multi-animal 3D social pose estimation, identification and behaviour embedding with a few-shot learning framework	113
The TRIPOD-P reporting guideline for improving the integrity and transparency of predictive analytics in healthcare through study protocols	111
Seeking a quantum advantage for machine learning	111
Collaborative creativity in AI	110
A personalized time-resolved 3D mesh generative model for unveiling normal heart dynamics	110
Deciphering RNA–ligand binding specificity with GerNA-Bind	109
Learning high-level visual representations from a child’s perspective without strong inductive biases	108
Neural scaling of deep chemical models	107
Deep learning supported discovery of biomarkers for clinical prognosis of liver cancer	106
Bridging peptide presentation and T cell recognition with multi-task learning	105
Foundation models in healthcare require rethinking reliability	100
Accurate and robust protein sequence design with CarbonDesign	100
A multi-modal pre-training transformer for universal transfer learning in metal–organic frameworks	99
Image-based generation for molecule design with SketchMol	94
Multimodal learning with graphs	94
Human–AI adaptive dynamics drives the emergence of information cocoons	92
Life-threatening ventricular arrhythmia detection challenge in implantable cardioverter–defibrillators	91
Unifying multi-sample network inference from prior knowledge and omics data with CORNETO	89
Morphological flexibility in robotic systems through physical polygon meshing	89
What is in your LLM-based framework?	88
Efficient generation of protein pockets with PocketGen	87
LLM-based agentic systems in medicine and healthcare	87

Unsupervised learning of topological non-Abelian braiding in non-Hermitian bands	86
Tandem mass spectrum prediction for small molecules using graph transformers	86
Algorithmic fairness and bias mitigation for clinical machine learning with deep reinforcement learning	85
Learning plastic matching of robot dynamics in closed-loop central pattern generators	85
Uncertainty-guided dual-views for semi-supervised volumetric medical image segmentation	85
Advanced AI assistants that act on our behalf may not be ethically or legally feasible	85
ResGen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling	84
Improving de novo molecular design with curriculum learning	84
Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering	83
Writing the rules in AI-assisted writing	82
Anniversary AI reflections	82
ARNLE model identifies prevalence potential of SARS-CoV-2 variants	82
Enhancing deep learning-based field reconstruction with a differentiable learning framework	81
A multimodal cell-free RNA language model for liquid biopsy applications	81
Reusability report: Exploring the transferability of self-supervised learning models from single-cell to spatial transcriptomics	80
Foundation models and the privatization of public knowledge	80
Reconstructing growth and dynamic trajectories from single-cell transcriptomics data	79
A method for multiple-sequence-alignment-free protein structure prediction using a protein language model	78
Geometric deep learning reveals the spatiotemporal features of microscopic motion	78
A neuro-vector-symbolic architecture for solving Raven’s progressive matrices	78
From attribution maps to human-understandable explanations through Concept Relevance Propagation	77
Synthetic data accelerates the development of generalizable learning-based algorithms for X-ray image analysis	76
Moving towards genome-wide data integration for patient stratification with Integrate Any Omics	75
Differentiable visual computing for inverse problems and machine learning	74
Human-behaviour-based social locomotion model improves the humanization of social robots	74
Model-based reinforcement learning for ultrasound-driven autonomous microrobots	74
Publisher Correction: A neural machine code and programming framework for the reservoir computer	73
A social network for AI	73
Distinguishing two features of accountability for AI technologies	72
An interaction-derived graph learning framework for scoring protein–peptide complexes	71
Mask-prior-guided denoising diffusion improves inverse protein folding	71
Mode switching in organisms for solving explore-versus-exploit problems	71
Lessons from a challenge on forecasting epileptic seizures from non-cerebral signals	71
Active learning for optimal intervention design in causal models	70
Learning integral operators via neural integral equations	69
Benchmarking AI-powered docking methods from the perspective of virtual screening	69
Closed-form continuous-time neural networks	68
Sampling-enabled scalable manifold learning unveils the discriminative cluster structure of high-dimensional data	68
A collaborative constrained graph diffusion model for the generation of realistic synthetic molecules	68
Successful implementation of the EU AI Act requires interdisciplinary efforts	68
Artificial intelligence-powered electronic skin	67
Defending ChatGPT against jailbreak attack via self-reminders	67
Learning intermediate physical states for inverse metasurface design	67
CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling	66
Interpretable meta-score for model performance	66
Delineating the effective use of self-supervised learning in single-cell genomics	66
Advances, challenges and opportunities in creating data for trustworthy AI	66
Leveraging language model for advanced multiproperty molecular optimization via prompt engineering	66
On board with COMET to improve omics prediction models	65
Current-diffusion model for metasurface structure discoveries with spatial-frequency dynamics	64
A unified deep framework for peptide–major histocompatibility complex–T cell receptor binding prediction	64
Machine learning-enabled globally guaranteed evolutionary computation	64
A new eye on inherited retinal disease	62
Incorporating physics into data-driven computer vision	62
Listening in to perceived speech with contrastive learning	62
Invalid SMILES are beneficial rather than detrimental to chemical language models	61
A generalizable deep learning framework for inferring fine-scale germline mutation rate maps	61
Versatile cardiovascular signal generation with a unified diffusion transformer	61
Fast, scale-adaptive and uncertainty-aware downscaling of Earth system model fields with generative machine learning	61
A computational framework for neural network-based variational Monte Carlo with Forward Laplacian	60
Solving sparse finite element problems on neuromorphic hardware	60
Large language models challenge the future of higher education	60
The incentive gap in data work in the era of large models	59
Deep learning for predicting rate-induced tipping	58
Teaching machines to blend electrolyte cocktails	58
Predicting the conformational flexibility of antibody and T cell receptor complementarity-determining regions	56
Data-driven discovery of movement-linked heterogeneity in neurodegenerative diseases	56
Lossless data compression by large models	56
Three types of incremental learning	55
Unsupervised ensemble-based phenotyping enhances discoverability of genes related to left-ventricular morphology	55
Augmenting large language models with chemistry tools	55
Deep transfer operator learning for partial differential equations under conditional shift	55
Mitigating the missing-fragmentation problem in de novo peptide sequencing with a two-stage graph-based deep learning model	55
AI podcasts for the summer	54
The importance of negative training data for robust antibody binding prediction	54
Competing Biases underlie Overconfidence and Underconfidence in LLMs	53
Generalized biological foundation model with unified nucleic acid and protein language	52
A process-centric manipulation taxonomy for the organization, classification and synthesis of tactile robot skills	52
A family of large language models for materials research with insights into model adaptability in continued pretraining	52
Deconstructing the generalization gap	52

When large language models are reliable for judging empathic communication	51
Deep learning-based prediction of the selection factors for quantifying selection in immune receptor repertoires	51
Labelling instructions matter in biomedical image analysis	50
Design of prime-editing guide RNAs with deep transfer learning	50
Discovering neural policies to drive behaviour by integrating deep reinforcement learning agents with biological neural networks	50
Realistic morphology-preserving generative modelling of the brain	49
Synergy-based robotic quadruped leveraging passivity for natural intelligence and behavioural diversity	49
Space missions out of this world with AI	49
Predicting the prevalence of complex genetic diseases from individual genotype profiles using capsule networks	49
Publisher Correction: Advancing ethics review practices in AI research	48
Aligning generalization between humans and machines	47
Automated construction of cognitive maps with visual predictive coding	47
Visual speech recognition for multiple languages in the wild	46
Pan-Peptide Meta Learning for T-cell receptor–antigen binding recognition	46
Multiscale topology-enabled structure-to-sequence transformer for protein–ligand interaction predictions	46
An integrated framework to accelerate protein design through mutagenesis	44
Reply to: Inability of a graph neural network heuristic to outperform greedy algorithms in solving combinatorial optimization problems	44
Author Correction: Predicting equilibrium distributions for molecular systems with deep learning	43
On the caveats of AI autophagy	43
Type II mechanoreceptors and cuneate spiking neuronal network enable touch localization on a large-area e-skin	43
A large-scale randomized study of large language model feedback in peer review	43
Geometric deep learning of particle motion by MAGIK	42
Towards a universal model for spin–orbit physics	42
A framework for tool cognition in robots without prior tool learning or observation	42
Towards unveiling sensitive and decisive patterns in explainable AI with a case study in geometric deep learning	42
A disease-specific language model for variant pathogenicity in cardiac and regulatory genomics	41
Why design choices matter in recommender systems	41
Weak signal extraction enabled by deep neural network denoising of diffraction data	41
Accelerating protein engineering with fitness landscape modelling and reinforcement learning	41
Clinical large language models with misplaced focus	40
A ‘programming’ framework for recurrent neural networks	40
Embodied large language models enable robots to complete complex tasks in unpredictable environments	40
An adaptive graph learning method for automated molecular interactions and properties predictions	40
Parameter-efficient fine-tuning of large-scale pre-trained language models	40
Learning motif-based graphs for drug–drug interaction prediction via local–global self-attention	40
Efficient rare event sampling with unsupervised normalizing flows	40