OpenAI Blog
Feb 13
GPT-5.2 derives a new result in theoretical physics
A new preprint shows GPT-5. 2 proposing a new formula for a gluon amplitude, later formally proved and verified by OpenAI and academic collaborators.
Extending single-minus amplitudes to gravitons
A new preprint extends single-minus amplitudes to gravitons, with GPT-5. 2 Pro helping derive and verify nonzero graviton tree amplitudes in quantum gravity.
Related stories
arXiv Machine Learning
Sep 18
Deep Learning Detection of Beyond-General-Relativity Deviations in Gravitational-Wave Signals: A Detection-Threshold Study with Real LIGO Noise
The paper investigates machine‑learning detection of controlled deviations from General Relativity in gravitational‑wave signals. Using a hybrid classifier that combines a one‑dimensional convolutional neural network with ten hand‑crafted waveform statistics, the authors train on General‑Relativistic and modified waveforms and test on a deviation type not seen during training. They find a detection threshold at a dimensionless strength coefficient β ≈ 0.25 when using real GW150914 strain and real H1 detector noise, with accuracy improving from chance at β ≤ 0.2 to perfect classification at β ≥ 0.5.
arXiv AI
Jun 30
Gravitational Duals from Equations of State II: Large Hierarchies and False Vacua
arXiv:2606. 30117v1 Announce Type: cross Abstract: We investigate the reconstruction of holographic duals for strongly coupled quantum field theories in regimes characterized by large hierarchies and the presence of false vacua.
DeepMind Blog
May 17
Introducing Google Antigravity 2.0
arXiv AI
Sep 23
gwBenchmarks: Stress-Testing LLM Agents on High-Precision Gravitational Wave Astronomy
arXiv:2605.11269v2 Announce Type: replace-cross Abstract: Modern gravitational wave astronomy relies on modeling tasks that often require months of graduate-level effort, including building fast wave...
DeepMind Blog
Nov 18, 2025
Introducing Google Antigravity
arXiv Machine Learning
Sep 15
Approximating neutron-star radii using gravitational-wave only measurements with symbolic regression
The paper presents a symbolic regression approach that derives an approximate expression for neutron‑star radii using only gravitational‑wave data from binary inspirals. By training on TOV solutions with piecewise polytropic equations of state, the authors obtain a formula that reproduces radii with differences of only a few hundred meters across a wide range of neutron‑star parameters. The method is validated on realistic dense‑matter EOSs and applied to the GW170817 event, comparing the inferred radii to existing distributions.
arXiv Machine Learning
Jun 9
Integral Formulas for Vector Signal Tensor Products
arXiv:2603. 08630v2 Announce Type: replace Abstract: We derive integral formulas that simplify the Vector Signal Tensor Product recently introduced by Xie et al.
arXiv Machine Learning
Jun 16
Learning the generating functional for variance reduction in lattice QCD
arXiv:2606. 15986v1 Announce Type: cross Abstract: The generating functional in quantum field theory provides the natural framework for constructing correlation functions as derivatives with respect to source operators.
arXiv Machine Learning
Jun 15
Binary Black Hole Parameter Estimation with Hybrid CNN-Transformer Neural Networks
arXiv:2606. 13941v1 Announce Type: cross Abstract: The detection of gravitational waves has revolutionized our ability to explore fundamental aspects of the Universe.
arXiv Machine Learning
Jun 10
Efficient AI-Inspired Reduction of Feynman Integrals via Tube Seeding
arXiv:2606. 10698v1 Announce Type: cross Abstract: In this paper, we use machine learning to discover a new seeding strategy for integration-by-parts reduction of Feynman integrals, which is a frequent bottleneck in state-of-the-art calculations in theoretical particle and gravitational-wave physics.
arXiv Machine Learning
Jun 25
Flexible Gravitational-Wave Parameter Estimation with Transformers
arXiv:2512. 02968v2 Announce Type: replace-cross Abstract: Gravitational-wave data analysis relies on accurate and efficient methods to extract physical information from noisy detector signals, yet the increasing rate and complexity of observations represent a growing challenge.