Training effective foundation models requires massive and organized datasets, yet scientific domains such as nuclear fusion present unique challenges due to largely heterogeneous and sparse data. Here we characterize the data used in developing such a model: with over 20 sensor types spanning 5 orders of magnitude in sampling rate, mixed tensor structures (point measurements, spectrograms, images), and nonstationary physics. We analyze our input complexity and discuss trade-offs between temporal context and frequency resolution. Our analysis provides a template for representing multi-modal fluctuation data at scale, with implications for both multi-modal control systems and nuclear fusion.
Significant accuracy improvements are achieved across seven observables, including charge radii, masses, binding energies, separation energies, and decay energies, with the training loss decreasing by over 98% across all tasks.
Shi-Jie Guo, Shou-Yu Wang, E.-H. Wang et al.· Chinese Physics Letters· 0 citations
The first application of sparse-autoencoder-based mechanistic interpretability to particle physics suggests that mechanistic interpretability can reveal learned latent physics encoded within a model's internal representation and help design downstream tasks that exploit it.
Raphaël Bonnet-Guerrini, Johann Ioannou-Nikolaides, I. Timiryasov et al.· 0 citations
Bayesian sampling methods in very-long-baseline interferometry enable reliable imaging of astronomical sources with a record high angular resolution. However, even after the transition from classical MCMC methods to flow-based neural networks, they remain computationally expensive as the number of recovered features...
L. E. Kats, A. Goldin, A. Andrianov· Monthly notices of the Royal...· 0 citations
The dipole model provides a powerful framework for describing high-energy nuclear interactions, particularly in the regime of dense gluonic matter. However, accurately evolving the dipole--nucleus scattering amplitude remains a major computational challenge because it is governed by nonlinear QCD evolution equations. T...
Junaid S. Khan, Rebecca L. Lustberg, F. Olness et al.· 0 citations
MUSHY is an amortized Bayesian inference method, using a rectified flow model with multimodal inputs to learn a posterior distribution over physical parameters, that can precisely infer physical parameters with well-calibrated posteriors from any combination of surveys.
This work introduces S-matrix informed neural networks (SINNs), and demonstrates their ability to learn scattering amplitudes directly from data while respecting first principles, and develops a novel data selection procedure which uses the response of constrained neural network ensembles to identify a set of experimen...
W. Smith, A. Rodas, Marius D. Thomas et al.· 0 citations
Research Manager Mike Walker reflects on a career spanning hacker culture, cybersecurity, and AI, sharing lessons about trust, teamwork, leadership, and why solving difficult problems requires curiosity, judgment, agility—and an understanding of the rules. The post Learn the rules: The hacker mindset appeared first on Microsoft Research.
Learn how Google Antigravity and our Cheminformatics Skills accelerate drug discovery workflows, search compounds, and propose novel rigidified macrocycles.
MIT News · Artificial Intelligence· news.mit.eduOct 28, 2026
Professor Paris Smaragdis develops computers with superhuman hearing for applications like speech recognition, music understanding, and environmental monitoring.
MIT News · Artificial Intelligence· news.mit.eduOct 27, 2026
Prabha Jha used her experience with MIT Open Learning’s program for senior executives to develop AI solutions for her employer, Verizon, and to write a book.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
