A domain-specific multi-agent AI framework for astrophysical research that combines literature retrieval, multiwavelength data access and analysis, theoretical modeling, and research ideation within a unified environment, with its current implementation focused on blazar research.
See LICENSE fileAstrophysicsMulti-Agent AIBlazar Research
This paper presents a comparative analysis of two neural network models for Initial Orbit Determination from Very Short Arcs, addressing challenges in Near-Earth Object surveillance where classical methods fail due to insufficient observational arcs.
See LICENSE fileInitial Orbit DeterminationNear-Earth ObjectsAsteroid Surveillance
This paper presents a bonded-particle discrete-element approach with self-gravity for studying collisional and gravitational evolution, fragmentation, and reaccumulation of small rubble-pile asteroids.
See LICENSE fileasteroid dynamicsdiscrete element methodgranular media
Proposes a continuous delayed-memory stochastic gradient descent method combined with neural network-parameterized stochastic differential equations to model quasar brightness variations from astronomical survey time series.
See LICENSE filequasartime seriesstochastic differential equations
This case study applies Bayesian inference to radio interferometry, using sparse satellite sampling to infer 137 MHz zenith-pointed power beams of Murchison Widefield Array receiving elements and quantify deviations from nominal models.
See LICENSE fileradio astronomybeam modelingBayesian inference
PHASE uses reusable solution operators and residual error correction to model multiscale incompressible magnetohydrodynamics across varying parameter settings, supporting fast inference for plasma dynamics.
See LICENSE filemagnetohydrodynamicssurrogate modelingmulti-regime modeling
Physics-informed neural networks recover the time-time component of static, spherically symmetric black-hole exterior metrics from a reduced ODE, using residual losses and asymptotic boundary constraints across charge and cosmological-constant sweeps.
See LICENSE filephysics-informed neural networksblack-hole exterior metricsgeneral relativity
ROBIN-PIP targets robust Bayesian field-level inference for galaxy surveys at nonlinear scales, combining physics-informed priors with more expressive representations of the galaxy bias relation.
See LICENSE fileBayesian field-level inferencephysics-informed priorsgalaxy bias
This astrophysics paper uses neural-network emulation to model synchrotron self-Compton emission from very high-energy gamma-ray burst afterglows, while Bayesian inference addresses the computational cost of parameter estimation.
See LICENSE filegamma-ray burstshigh-energy afterglowsneural-network emulation