AI4Science field
Research releases and public discussion in this scientific field, ordered by time.
Nature Computational Science, Published online: 09 October 2026; doi:10.1038/s43588-026-01071-0 A physics-based in situ learning method is demonstrated that directly measures optical gradients with photonic integrated circuits, allowing diverse circuit topologies to be trained after fabrication. This hardware achieves matrix design, multiwavelength computation and rapid meta-learning without requiring a full digital
A few days ago OpenAI released a paper generated by "an internal frontier model" of Navier-Stokes fame titled A Coulomb ground-state density without Kohn-Sham ensemble representation . The model considers a system with 2 nuclei and 3 electrons. It then spends 40 pages (seemingly) proving that for "sufficiently large" internuclear distance, restricted open-shell ensemble KS-DFT cannot describe the ground-state density
Automated building design must comply with seismic and wind codes and satisfy structural mechanics constraints, yet most existing agents produce visually plausible models without verification grounded in mechanical analysis and code compliance. We introduce the Vibe Building task and propose PE-Loop (Physics-Engine-in-the-Loop), an agent in which a deterministic physics engine is the sole source of evaluation signals
At the International Conference on Machine Learning (ICML 2026) , Dezhi Luo, Yijiang Li, Maijunxian Wang, Tianwei Zhao, Bingyang Wang, Siheng Wang, Pinyuan Feng, Pooyan Rahmanzadehgervi, Ziqiao Ma , and Hokin Deng presented their work Vision Language Models Cannot Reason About Physical Transformation . In this interview, William Yijiang Li tells us more about the research, the method the team used, and the controlled
Physics-based neuromorphic computing promises major gains in performance and energy efficiency, but training remains largely in silico, requiring costly physical modeling and suffering from fabrication-induced errors. Here we introduce on-chip, in situ physical gradient descent (INSPIRE), a general training method for photonic integrated circuits. INSPIRE uses on-chip synthetic time-reversal holography to measure the
Partnered human-humanoid interaction couples locomotion with continuous physical contact. A humanoid needs to coordinate with a person's motion while responding to interaction forces and maintaining stable and natural movement. We present CoDance, a framework for learning reactive and compliant human-humanoid interaction from video. We study partnered dancing as a challenging instantiation, where a humanoid coordinat
Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do so effectively, these models must not only generate visually appealing frames, but also predict how environments dynamically evolve when executing goal-directed actions. While evaluating these capabilities is crucial, existing benchmarks focus mainly on single short actions or step-by-step instruction
AI Models & Platforms Anthropic Says Claude Computed a Nine-Loop Particle Physics Amplitude Add Unite.AI to your preferred sources on Google Anthropic on September 25, 2026 published a guest post on its research site reporting that two of its physicists used the company’s Claude model to compute the six-particle scattering amplitude in planar N=4 super Yang-Mills at nine loops, answering a public challenge issued to
Science In this guest post, physicist and science writer Matt von Hippel shares what happened when he issued a challenge to AI companies regarding a problem in his former subfield of theoretical physics. It’s not often that you issue a challenge, only to see it beaten a month later. But we’re living in unusual times. Let me introduce myself: I’m Matt von Hippel. I used to be a theoretical physicist; these days I’m a
AI agents are beginning to automate research and development across the AI stack, from improving training efficiency to optimizing inference. A natural next step is to improve the research efficiency of the agents themselves. When an AI research agent's own code is the object of optimization, each accepted rewrite becomes the agent that the next round edits. We refer to this loop as recursive self-improvement. Its si
Objectives in scientific machine learning are often prescribed as a sum of several terms, such as the residual, boundary, initial, and data losses of a physics-informed neural network. In the pullback-corrected scalar auxiliary variable (PB--SAV) method, one scalar tracks the shifted objective while the component gradients build a positive semidefinite curvature correction of rank at most the number of components. We
The rapid progress of artificial intelligence is reshaping robotics and accelerating the adoption of learning-based approaches. While purely data-driven methods have achieved remarkable success in computer vision and natural language processing, robotics remains constrained by limited data, complex real-world interactions, and the need for reliable operation. These challenges have motivated the exploration of physics
Reacting to sudden physical hazards (catching a slipping plate, dodging a falling knife) is both a meaningful test of embodied intelligence and a hard requirement for deploying multimodal large language models (MLLMs) as the decision coreof household robots. Existing evaluations, however, probe intuitive physics passively through question answering over videos, or target deliberate, long-horizon tasks such as navigat
We propose Puffin-World, a unified multimodal architecture that integrates physical understanding, spatial simulation, and 3D world generation and reconstruction without relying on external offline modules. To reliably construct and interact with 3D worlds, our framework jointly models three native world states: physics (gravity field and latitude), geometry (depth), and appearance (image), together with a unified Om
Evaluating physical reasoning in video models is difficult because absolute motion measurements depend on frame rate, object scale, and camera calibration, all of which are often ambiguous or unavailable in generated video. We propose a different approach. When two objects in the same scene obey the same physical law, their motions must satisfy predictable relationships, and these relationships hold independent of ca
Progress in physics has long been driven by ingenious experiments conceived by human experts. Recently, design methods driven by artificial intelligence (AI) have begun to move beyond tuning a handful of parameters to proposing entirely new experimental layouts. The discovered configurations often challenge established design conventions while matching or even exceeding the performance of human-designed set-ups. We f
This paper introduces a framework that integrates the dynamic stiffness matrix (DSM) with physics-informed neural networks (PINN). The DSM-PINN embeds physical constraints within the model and demonstrates robustness, particularly when addressing limited datasets across diverse investigations. In this approach, deep neural network outputs approximate the displacement fields of element nodes. Unlike the finite element
Equivariant graph neural networks provide a principled way to model geometric systems, but efficient first-order architectures remain limited in how vector information can be transformed as it moves across a graph. We introduce \textsc{ESNN}, an Equivariant Sheaf Neural Network that enriches this interaction by learning directed, matrix-valued transport between neighboring vector features while preserving exact Eucli
In many computational science and engineering problems, repeatedly solving fully resolved physics-based models to design for a quantity of interest (QoI) can quickly become intractable, requiring the use of low-fidelity models to predict the same QoI but introduce errors where some features are neglected or are otherwise inaccurately resolved. We use Gaussian Functional Regression (GFR) to learn a correction to a 2D
Machine learning interatomic potentials (MLIPs) have become emerging tools in molecular modeling and computational chemistry. By learning high-dimensional potential energy surfaces from quantum chemical data, MLIPs enable accurate and efficient predictions of structural, thermodynamic, and dynamical properties. However, such models have limitations in predictions of electronic properties and the effects of static ele
AUG. 27, 2026 Astroparticle physics sits at the exciting intersection of astrophysics and particle physics and studies cosmic messengers: particles such as atomic nuclei, electrons, high-energy photons (light), and neutrinos accelerated in distant, cosmic phenomena and objects such as black holes, neutron stars, active galaxies, and supernovas, as well as other events from the early formation of the universe. By stud
Rapid and accurate fault detection in high-voltage transmission networks is essential for grid reliability and equipment protection. Transmission fault datasets are frequently imbalanced, and certain fault types produce electrical signatures that fall within the normal operating envelope, causing single-model classifiers to fail on safety-critical cases. This paper proposes a hybrid two-stage machine learning pipelin
We demonstrate an iterative approach to modeling a sparsely measured magnetic field in a large-bore solenoid. This approach uses a hybrid of traditional and machine learning techniques. The traditional technique is a linear least-squares fit using a series solution to Laplace's equation, while the machine learning technique involves the training of a physics-informed neural network (PINN) on the least-squares fit res
To achieve net-zero targets by 2050, it is critical to diversify renewable energy. Hydropower, wind, and solar energy dominate; geothermal energy remains underutilized. Conventional Enhanced Geothermal Systems (EGS) rely on hydraulic stimulation, which poses risks such as induced seismicity. To address this, Closed-Loop Geothermal Systems (CLGS) circulate working fluids in sealed tubing to avoid direct reservoir cont
Neural PDE operators are increasingly trained on reusable solver archives, yet validation often relies on clean prediction error and parameter-agnostic plausibility checks. We introduce cross-parameter relinking, a data-poisoning primitive that makes a triggered input select a valid solution from the same PDE family under an incorrect physical parameter. We term this a wrong-physics backdoor: the output remains physi
Authentication of edible oils in processed foods is important for food quality, fraud prevention, and regulatory compliance. This study establishes an integrated Raman spectroscopy and machine-learning framework that links intrinsic spectral organization, interpretable classification, and Physics-Informed Artificial Intelligence (PI-AI). Five edible oils were investigated in pure form and within a fried-potato-chip m
Many modern machine learning models can be understood as kernel-based function-space models, including Gaussian processes and neural tangent kernels. In scientific machine learning, differential operators are increasingly used to encode physical structure directly into such models. However, it has remained unclear under which conditions these constructions are valid machine learning models, i.e. preserve positive sem
New research reveals that current world models like Sora or Genie only simulate physics, ignoring what people think, want, or feel. The new "Mental World Modeling" framework adds mental variables such as beliefs and intentions. Even weaker language models using this approach outperform stronger models without mental modeling. The biggest bottleneck remains predicting how physical and mental states change together.
Researchers increasingly use AI to measure variables for downstream analyses, but ignoring prediction errors—even with accuracy above 90%—can cause substantial bias and invalid confidence intervals. Existing remedies like design-based supervised learning and prediction-powered inference rely on costly gold-standard labels. This paper introduces debiased inference with multiple imperfect measurements (DMM), a framewor
This work proposes a new interpretable neural operator framework, termed the Physics Informed Kernel Function Neural Operator (PIKFNO), which explicitly incorporates physics informed kernel functions derived from governing equations into the neural operator architecture. Unlike traditional neural operators such as DeepONet, which rely on deep networks to implicitly learn basis functions, PIKFNO constrains the trunk n
Physics-informed neural networks (PINNs) offer a flexible framework for forward and inverse problems, but their direct use in structural dynamics is limited by high dimensionality and model-form errors. Existing PINN-ROM couplings often rely on fixed reduced subspaces or assume complete physics. This work proposes a Reduced-Order Physics-Informed Neural Network (RO-PINN) with adaptive basis refinement, embedding redu
This article investigates physics-informed and hybrid machine learning strategies that incorporate physics knowledge into experimental data-driven deep-learning models for predicting bond quality and porosity of fused filament fabrication (FFF) parts. Three strategies are explored: incorporating physics constraints into the loss function, using physics model outputs as additional inputs, and pre-training with physics
We address the inverse problem of reactive transport in heterogeneous porous media, where unknown model parameters must be inferred from sparse experimental observations. The problem is complicated by strong nonlinearities, spatial heterogeneity, and limited data availability. We propose a Conditioned-on-Design Bayesian Physics-Informed Neural Network (CoDe-BPINN), which combines a domain-decomposed PINN solver with
This paper investigates physics-informed machine learning strategies for global sensitivity analysis (GSA) when both a physics-based model and experimental observations are available. Two machine learning techniques, deep neural networks (DNN) and Gaussian process (GP) modeling, are considered, and two strategies for incorporating physics knowledge are explored: enforcing physics constraints via loss functions, and p
We propose a methodology based on the standard ReLU Deep Neural Networks (DNN) to make predictions and quantify their uncertainty. Classically, people rely on linear, non-linear, or non-parametric kernel methods to fit and then predict the time series. As the universal approximation ability was revealed for DNN, its application has become more and more popular for prediction tasks in various scientific areas. However
Bayesian optimization (BO) is a sample-efficient, surrogate-based approach to black-box optimization (BBO), but its evaluation remains dominated by synthetic functions and hyperparameter optimization (HPO) tasks that are typically low-dimensional and single-objective. Engineering design poses a substantially different regime: problems are physics-based, often high-dimensional, constrained by requirements such as cost
This work presents a machine learning-symbolic regression (ML-SR) strategy to develop a physically interpretable formula for predicting low pressure CO2 adsorption capacity in hypothetical metal-organic frameworks (hMOFs). Four ML models were trained on a small dataset of 1,000 samples, and five key descriptors—largest cavity diameter, pore limiting diameter, void fraction, gravimetric surface area, and number of hyd
In mixed traffic, decision-making for autonomous vehicles faces three interrelated challenges: physics-based priors fail to capture latent interactive intentions and diverse driver behaviors; abrupt maneuvers cause non-stationarity, leaving long-tail safety events under-explored; and hybrid action spaces destabilize unified RL training due to temporal scale differences between continuous car-following and discrete la
Machine learning is rapidly reshaping constitutive modeling, offers new ways to learn material behavior directly from experimental data, and challenges long-established modeling paradigms. But with a growing number of machine-learning-based approaches available, how do they compare in practice? In this paper, we use the classic experimental data of Treloar to benchmark popular frameworks for hyperelasticity: (General
Natural language offers robots a flexible task interface, but target ambiguity in embodied environments arises not only from user intent; it can also result from missing task-relevant physical evidence in the current observation. Existing interactive disambiguation methods primarily obtain additional information by asking the user, whereas occlusion, restricted viewpoints, unreadable text, and unobserved targets requ
A new deep learning method enables molecular dynamics simulations over longer time scales while still achieving accurate physical property prediction. Published online in Nature Machine Intelligence on August 5, 2026, the method promises to overcome the femtosecond ceiling that limits conventional simulation time steps, potentially expanding the applicability of molecular simulations in fields such as materials scien
Traditional robot learning relies on collecting demonstrations and evaluating policies on real robots, which is slow, expensive, and hard to reproduce. We propose an Interactive World Simulator: a learned, action-conditioned video prediction model that predicts subsequent pixel frames given the current image and robot actions, without a physics engine. On a single RTX 4090, it runs for over 10 minutes at 15 FPS, prod
Many problems in machine learning and the sciences come down to the same task: you have a collection of data points and want to recover the distribution they came from—which values are common, and which are rare. Pinning down that distribution means estimating two quantities: the distribution’s density and, more useful as dimensionality grows, its score. The density is the smooth version of a histogram—high where poi
Scientific progress is often limited not by a lack of new ideas, but by the cost and complexity of testing them. New solutions are needed to make that testing easier—and researchers like Anima Anandkumar are leading the way. She has conducted cutting-edge research across academia and industry for over a decade, pioneering new AI algorithms that simulate physical systems with unprecedented speed and accuracy—in some c
Science Mar 23, 2026 Can AI do theoretical physics? In this guest post, professor of physics Matthew Schwartz decided to find out by supervising Claude through a real research calculation, start to finish, without ever touching a file himself. His account of what happened is below. - I guided Claude Opus 4.5 through a real theoretical physics calculation, encapsulating the complexity of code and computations behind t
Molecular Hessians, the second derivatives of the potential energy, are fundamental to many workflows in computational chemistry. Usually, accurate Hessians are computationally expensive to calculate and scale poorly with system size, whether computed using quantum chemistry methods or machine-learning interatomic potentials (MLIPs). In this work, we introduce Hessian interatomic potentials (HIPs), a deep learning mo