OOIR: Observatory of International Research

Papers

(The H4-Index of Nature Machine Intelligence is 82. The table below lists those papers that are above that threshold based on CrossRef citation counts [max. 250 papers]. The publications cover those that have been published in the past four years, i.e., from 2022-05-01 to 2026-05-01.)

Article	Citations
Physical benchmarks for testing algorithms	842
A challenge for the law and artificial intelligence	717
Towards reproducible robotics research	548
Author Correction: A 5′ UTR language model for decoding untranslated regions of mRNA and function predictions	543
Author Correction: Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	493
Investigating machine moral judgement through the Delphi experiment	460
Physically constrained generative adversarial networks for improving precipitation fields from Earth system models	421
A domain-adapted large language model to support clinicians in psychiatric clinical practice	380
Wing-strain-based flight control of flapping-wing drones through reinforcement learning	380
From embodied intelligence to physical AI	373
A soft-packaged and portable rehabilitation glove capable of closed-loop fine motor skills	333
A statistical mechanics framework for Bayesian deep neural networks beyond the infinite-width limit	305
Tailored structured peptide design with a key-cutting machine approach	300
A multi-modal deep language model for contaminant removal from metagenome-assembled genomes	240
Identifying spatial single-cell-level interactions with graph transformer	219
Materiality and risk in the age of pervasive AI sensors	211
Discussions of machine versus living intelligence need more clarity	197
Artificial intelligence tackles the nature–nurture debate	196
Reusability report: Deep learning-based analysis of images and spectroscopy data with AtomAI	188
Quantum circuit optimization with AlphaTensor	182
Maximum diffusion reinforcement learning	178
Robust virtual staining of landmark organelles with Cytoland	176
Are neural network representations universal or idiosyncratic?	174
Large language models still struggle with false beliefs	166
Fast and generalizable micromagnetic simulation with deep neural nets	155

A Global South perspective for ethical algorithms and the State	148
Recurrent graph optimal transport for learning 3D flow motion in particle tracking	148
Bringing artificial intelligence to business management	145
Reshaping the discovery of self-assembling peptides with generative AI guided by hybrid deep learning	143
The curious case of the test set AUROC	143
Next-generation phenotyping of inherited retinal diseases from multimodal imaging with Eye2Gene	142
Generative AI for designing and validating easily synthesizable and structurally novel antibiotics	139
Zero-shot transfer of protein sequence likelihood models to thermostability prediction	138
Transformer-based protein generation with regularized latent space optimization	137
Direct conformational sampling from peptide energy landscapes through hypernetwork-conditioned diffusion	135
Deep neural networks with controlled variable selection for the identification of putative causal genetic variants	134
Pseudodata-based molecular structure generator to reveal unknown chemicals	134
Inferring spatial single-cell-level interactions through interpreting cell state and niche correlations learned by self-supervised graph transformer	131
Advancing ethics review practices in AI research	130
A question of trust for AI research in medicine	130
AI pioneers win 2024 Nobel prizes	129
Error-controlled non-additive interaction discovery in machine learning models	128
Integrated structure prediction of protein–protein docking with experimental restraints using ColabDock	128
What’s the next word in large language models?	127
How to break information cocoons	124
LLMs displaying less cognitive bias are not necessarily better decision makers	123
A new perspective on the simulation of stochastic problems in fluid mechanics with diffusion models	120
Machine learning prediction of enzyme optimum pH	116
Laplace neural operator for solving differential equations	116
Codon language embeddings provide strong signals for use in protein engineering	116
Learning from models beyond fine-tuning	115
Deep spectral component filtering as a foundation model for spectral analysis demonstrated in metabolic profiling	114
Multi-animal 3D social pose estimation, identification and behaviour embedding with a few-shot learning framework	111
Functional mimicry of Ruffini receptors with fibre Bragg gratings and deep neural networks enables a bio-inspired large-area tactile-sensitive skin	109
AI reality check	108
Life-threatening ventricular arrhythmia detection challenge in implantable cardioverter–defibrillators	107
Seeking a quantum advantage for machine learning	106
Collaborative creativity in AI	106
Bridging peptide presentation and T cell recognition with multi-task learning	106
What is in your LLM-based framework?	105
Accurate and robust protein sequence design with CarbonDesign	105
A personalized time-resolved 3D mesh generative model for unveiling normal heart dynamics	103
Morphological flexibility in robotic systems through physical polygon meshing	103
Foundation models in healthcare require rethinking reliability	100
The TRIPOD-P reporting guideline for improving the integrity and transparency of predictive analytics in healthcare through study protocols	99
Deciphering RNA–ligand binding specificity with GerNA-Bind	98
Efficient generation of protein pockets with PocketGen	98
Image-based generation for molecule design with SketchMol	97
Human–AI adaptive dynamics drives the emergence of information cocoons	95
Unsupervised learning of topological non-Abelian braiding in non-Hermitian bands	91
A multi-modal pre-training transformer for universal transfer learning in metal–organic frameworks	91
Neural scaling of deep chemical models	91
Autonomous environment-adaptive microrobot swarm navigation enabled by deep learning-based real-time distribution planning	89
Multimodal learning with graphs	87
Uncertainty-guided dual-views for semi-supervised volumetric medical image segmentation	86

Unifying multi-sample network inference from prior knowledge and omics data with CORNETO	86
Learning high-level visual representations from a child’s perspective without strong inductive biases	85
Algorithmic fairness and bias mitigation for clinical machine learning with deep reinforcement learning	85
Tandem mass spectrum prediction for small molecules using graph transformers	84
Deep learning supported discovery of biomarkers for clinical prognosis of liver cancer	84
LLM-based agentic systems in medicine and healthcare	84
Advanced AI assistants that act on our behalf may not be ethically or legally feasible	83
ARNLE model identifies prevalence potential of SARS-CoV-2 variants	82
Towards generalizable and interpretable three-dimensional tracking with inverse neural rendering	82
Improving de novo molecular design with curriculum learning	82