NumPyro delivers a JIT-compilable iterative NUTS sampler by composing Pyro effect handlers with JAX transformations, achieving faster performance than prior implementations.
hub Mixed citations
Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro
Mixed citation behavior. Most common role is background (60%).
46
Pith papers citing it
Background
60% of classified citations
abstract
NumPyro is a lightweight library that provides an alternate NumPy backend to the Pyro probabilistic programming language with the same modeling interface, language primitives and effect handling abstractions. Effect handlers allow Pyro's modeling API to be extended to NumPyro despite its being built atop a fundamentally different JAX-based functional backend. In this work, we demonstrate the power of composing Pyro's effect handlers with the program transformations that enable hardware acceleration, automatic differentiation, and vectorization in JAX. In particular, NumPyro provides an iterative formulation of the No-U-Turn Sampler (NUTS) that can be end-to-end JIT compiled, yielding an implementation that is much faster than existing alternatives in both the small and large dataset regimes.
hub tools
citation-role summary
background 3
method 2
citation-polarity summary
representative citing papers
A new queryable binary dataset combining cross-build diversity, temporal history, and CVE labels with linked metadata for vulnerability research.
Derives Wasserstein bounds and explicit hyperparameter tuning rules for annealed Langevin dynamics in compositional score-based SBI, proving Linhart et al. (2026) allows larger steps and fewer total steps than Geffner et al. (2023) in the Gaussian case.
Bayesian nonparametric model for dynamic sparse networks with overlapping communities via completely random measures and latent Markov processes.
Large longitudinal RCT finds high rates of following AI personal advice but no sustained well-being gains versus a hobbies control condition.
AMIGO is an end-to-end differentiable forward model of JWST AMI that corrects detector systematics to recover high-precision astrometry and detect close high-contrast companions.
Symbolic regression yields an emulator for the radial Fourier transform of the Sérsic profile that enables 2.5 times faster galaxy profile fitting with minimal accuracy loss.
The authors derive drift and diffusion constraints plus a parameterization that forces neural SDE solutions to remain inside compact polyhedral domains, yielding better forecasts on real EMA suicide-risk datasets than standard baselines.
Introduces HICOBIAN, a differentiable fuzzy hierarchical cosmic-web bias model using sigmoid gradients for smooth region transitions, enabling accurate Bayesian field-level reconstruction of primordial density fields validated by Fourier-space statistics.
Extends NPE to mixed discrete-continuous parameter spaces via a factorized inference network combining an autoregressive classifier and generative model, trained jointly to yield accurate calibrated posteriors.
Variational predictive resampling iteratively imputes data from a variational predictive to produce posterior samples that converge to the exact Bayesian posterior in Gaussian models where mean-field VI retains a gap.
A fully Bayesian pixel-based Doppler imaging framework uses Gaussian Process priors and Hamiltonian Monte Carlo to simultaneously infer surface maps and geometric parameters from spectral data.
ADELIA is the first AD-enabled INLA system that computes exact hyperparameter gradients via a structure-exploiting multi-GPU backward pass, delivering 4.2-7.9x per-gradient speedups and 5-8x better energy efficiency than finite differences on models with up to 1.9 million latent variables.
Massive black hole binary mergers produce orphaned low-frequency signals in PTA pulsar terms that can be stacked for archival multiband gravitational-wave detection.
A differentiable forward model and likelihood enable probabilistic inference over many spatial morphologies for the Galactic Center gamma-ray Excess using variational methods on GPUs.
AI4BayesCode generates validated modular stateful MCMC samplers from natural language Bayesian model descriptions via LLM translation, modular blocks, and recursive stateful composition.
Combined five-PTA dataset yields posterior on SGWB power-law amplitude and index consistent with nonzero signal but below 5-sigma significance, with reconstructed angular correlations matching the Hellings-Downs prediction.
OzDES final release delivers 62 new reverberation-mapped black hole masses and tighter lag-luminosity relations for Hβ, MgII, and CIV in high-redshift AGN after correcting for survey-length selection effects.
Photon counting readout detects weak postmerger gravitational wave signals at a rate of about 1 in 100 for SNR 0.2 and yields a twofold improvement in neutron star radius measurement after 20,000 events.
Dorito enables diffraction-limited image reconstruction from JWST AMI observations by deconvolving images or Fourier observables using maximum entropy and total variation regularization.
Conversational AI matches self-directed internet search in increasing belief in true political information and decreasing belief in misinformation.
RefineStat improves small language model performance on probabilistic program synthesis by adding semantic constraint enforcement and diagnostic-aware refinement, producing syntactically and statistically reliable code that often matches larger models.
BSA-TNP is a new neural process model with KRBlocks and biased scan attention that claims to match top accuracy while scaling inference to over 1M points in under a minute on a single GPU and supporting translation invariance.
Systematic search of 377k Gaia DR3 AGN light curves finds no reliable periodic SMBHB candidates after red-noise modeling and empirical false-alarm testing; all survivors lie in the few-cycle regime.
citing papers explorer
-
Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro
NumPyro delivers a JIT-compilable iterative NUTS sampler by composing Pyro effect handlers with JAX transformations, achieving faster performance than prior implementations.
-
ASSEMBLAGE-DEEPHISTORY: A Cross-Build Binary Dataset with Temporal Coverage
A new queryable binary dataset combining cross-build diversity, temporal history, and CVE labels with linked metadata for vulnerability research.
-
Theoretical guidelines for annealed Langevin dynamics in compositional simulation-based inference
Derives Wasserstein bounds and explicit hyperparameter tuning rules for annealed Langevin dynamics in compositional score-based SBI, proving Linhart et al. (2026) allows larger steps and fewer total steps than Geffner et al. (2023) in the Gaussian case.
-
Dynamic sparse graphs with overlapping communities
Bayesian nonparametric model for dynamic sparse networks with overlapping communities via completely random measures and latent Markov processes.
-
People readily follow personal advice from AI but it does not improve their well-being
Large longitudinal RCT finds high rates of following AI personal advice but no sustained well-being gains versus a hobbies control condition.
-
AMIGO: a Data-Driven Calibration of the JWST Interferometer
AMIGO is an end-to-end differentiable forward model of JWST AMI that corrects detector systematics to recover high-precision astrometry and detect close high-contrast companions.
-
Using Symbolic Regression to Emulate the Radial Fourier Transform of the S\'ersic profile for Fast, Accurate and Differentiable Galaxy Profile Fitting
Symbolic regression yields an emulator for the radial Fourier transform of the Sérsic profile that enables 2.5 times faster galaxy profile fitting with minimal accuracy loss.
-
Neural Stochastic Differential Equations on Compact State Spaces: Theory, Methods, and Application to Suicide Risk Modeling
The authors derive drift and diffusion constraints plus a parameterization that forces neural SDE solutions to remain inside compact polyhedral domains, yielding better forecasts on real EMA suicide-risk datasets than standard baselines.
-
Differentiable Fuzzy Cosmic-Web for Field Level Inference
Introduces HICOBIAN, a differentiable fuzzy hierarchical cosmic-web bias model using sigmoid gradients for smooth region transitions, enabling accurate Bayesian field-level reconstruction of primordial density fields validated by Fourier-space statistics.
-
Mixed neural posterior estimation for simulators with discrete and continuous parameters
Extends NPE to mixed discrete-continuous parameter spaces via a factorized inference network combining an autoregressive classifier and generative model, trained jointly to yield accurate calibrated posteriors.
-
Variational predictive resampling
Variational predictive resampling iteratively imputes data from a variational predictive to produce posterior samples that converge to the exact Bayesian posterior in Gaussian models where mean-field VI retains a gap.
-
Bayesian Doppler Imaging: Simultaneous Inference of Surface Maps and Geometric Parameters
A fully Bayesian pixel-based Doppler imaging framework uses Gaussian Process priors and Hamiltonian Monte Carlo to simultaneously infer surface maps and geometric parameters from spectral data.
-
ADELIA: Automatic Differentiation for Efficient Laplace Inference Approximations
ADELIA is the first AD-enabled INLA system that computes exact hyperparameter gradients via a structure-exploiting multi-GPU backward pass, delivering 4.2-7.9x per-gradient speedups and 5-8x better energy efficiency than finite differences on models with up to 1.9 million latent variables.
-
Archival Multiband Gravitational-Wave Signals from Massive Black Hole Binary Mergers
Massive black hole binary mergers produce orphaned low-frequency signals in PTA pulsar terms that can be stacked for archival multiband gravitational-wave detection.
-
High-dimensional inference for the $\gamma$-ray sky with differentiable programming
A differentiable forward model and likelihood enable probabilistic inference over many spatial morphologies for the Galactic Center gamma-ray Excess using variational methods on GPUs.
-
AI4BayesCode: From Natural Language Descriptions to Validated Modular Stateful Bayesian Samplers
AI4BayesCode generates validated modular stateful MCMC samplers from natural language Bayesian model descriptions via LLM translation, modular blocks, and recursive stateful composition.
-
Stochastic gravitational-wave background search using data from five pulsar timing arrays
Combined five-PTA dataset yields posterior on SGWB power-law amplitude and index consistent with nonzero signal but below 5-sigma significance, with reconstructed angular correlations matching the Hellings-Downs prediction.
-
OzDES Reverberation Mapping of Active Galactic Nuclei: Final Data Release, Black-Hole Mass Results, & Scaling Relations
OzDES final release delivers 62 new reverberation-mapped black hole masses and tighter lag-luminosity relations for Hβ, MgII, and CIV in high-redshift AGN after correcting for survey-length selection effects.
-
Photon counting readout for detection and inference of gravitational waves from neutron star merger remnants
Photon counting readout detects weak postmerger gravitational wave signals at a rate of about 1 in 100 for SNR 0.2 and yields a twofold improvement in neutron star radius measurement after 20,000 events.
-
Image reconstruction with the JWST Interferometer
Dorito enables diffraction-limited image reconstruction from JWST AMI observations by deconvolving images or Fourier observables using maximum entropy and total variation regularization.
-
Conversational AI increases political knowledge as effectively as self-directed internet search
Conversational AI matches self-directed internet search in increasing belief in true political information and decreasing belief in misinformation.
-
RefineStat: Efficient Exploration for Probabilistic Program Synthesis
RefineStat improves small language model performance on probabilistic program synthesis by adding semantic constraint enforcement and diagnostic-aware refinement, producing syntactically and statistically reliable code that often matches larger models.
-
Scalable Spatiotemporal Inference with Biased Scan Attention Transformer Neural Processes
BSA-TNP is a new neural process model with KRBlocks and biased scan attention that claims to match top accuracy while scaling inference to over 1M points in under a minute on a single GPU and supporting translation invariance.
-
A search for periodic AGN variability in $\textit{Gaia}$ Data Release 3
Systematic search of 377k Gaia DR3 AGN light curves finds no reliable periodic SMBHB candidates after red-noise modeling and empirical false-alarm testing; all survivors lie in the few-cycle regime.
-
A "Black Hole Star" Reveals the Remarkable Gas-Enshrouded Hearts of the Little Red Dots
A source 660 million years after the Big Bang is interpreted as a black hole star with a dust-free dense gas atmosphere, implying Little Red Dots have black hole masses overestimated by orders of magnitude.
-
Stories in Space: In-Context Learning Trajectories in Conceptual Belief Space
LLMs perform in-context learning as trajectories through a structured low-dimensional conceptual belief space, with the structure visible in both behavior and internal representations and causally manipulable via interventions.
-
A hierarchical Bayesian pipeline for soliton-plus-NFW inference on SPARC rotation curves: diagnostics and prior-boundary behaviour
A hierarchical Bayesian pipeline applied to 106 SPARC galaxies yields posteriors that reach prior boundaries for soliton parameters, indicating no detectable interior population-level soliton within the Schive-normalized model and standard priors.
-
What You Don't Know Won't Hurt You: Self-Consistent Hierarchical Inference with Unknown Follow-up Selection Strategies
Hierarchical Bayesian inference allows accurate recovery of intrinsic astrophysical source populations even when follow-up selection is unknown and correlated with parameters of interest.
-
Bayesian Modeling and Prediction of Generalized Contact Matrices
A Bayesian model for multi-feature contact matrices that uses tensor structures and contingency table theory to satisfy structural constraints and impute missing contact features, validated on simulations and US/German survey data.
-
Towards E-Value Based Stopping Rules for Bayesian Deep Ensembles
E-value sequential tests enable early stopping of MCMC sampling in Bayesian deep ensembles, often needing only a fraction of the full budget while improving over standard deep ensembles.
-
A unified harmonic framework for dark siren cosmology
The GW-galaxy cross-correlation method, unified with spectral sirens in a harmonic framework, can measure H0 to 1% and Omega_m to 5% precision with 2 years of data from next-generation detectors like Einstein Telescope and Cosmic Explorer.
-
Towards Understanding Sycophancy in Language Models
Sycophancy is prevalent in state-of-the-art AI assistants and is likely driven in part by human preferences that favor agreement over truthfulness.
-
Corrected Integrated Laplace Approximation for Bayesian Inference in Latent Gaussian Models
An importance sampling correction is added to integrated Laplace approximation so that the approximate posterior for latent Gaussian models converges to the true posterior as the number of samples grows.
-
Gravitational-wave constraints on $H_0$ are robust to (putative) redshift evolution in the binary black hole mass spectrum at current sensitivity
Spectral-siren H0 constraints from GWTC-4.0 binary black holes remain robust when the mass spectrum is permitted to evolve with redshift at current detector sensitivity.
-
Discovery of a compact hierarchical triple main-sequence star system while searching for binary stars with compact objects
A new compact hierarchical triple main-sequence star system G1010 was discovered through combined low- and high-SNR spectroscopy, Gaia DR3 data, and TESS light curve analysis, showing an inner eclipsing binary rather than a compact object companion.
-
The DESI DR1 Peculiar Velocity Survey: growth rate measurements from the maximum likelihood fields method
DESI DR1 peculiar velocity data yields fσ8(z_eff=0.07) = 0.450 ± 0.055, consistent with Planck ΛCDM and GR growth index γ = 0.58 ± 0.11.
-
Symbolic Emulators for Cosmology: Accelerating Cosmological Analyses Without Sacrificing Precision
Symbolic emulators approximate key Lambda CDM functions to 0.001-0.05% accuracy across relevant redshifts and Omega_m values, enabling faster 3x2pt inference with consistent results.
-
LITMUS: Bayesian Lag Recovery in Reverberation Mapping with Fast Differentiable Models
LITMUS introduces a differentiable Bayesian lag recovery framework that outperforms JAVELIN on OzDES-like mock data by reducing false positives from seasonal aliasing.
-
Temporal Point Process Modeling of Aggressive Behavior Onset in Psychiatric Inpatient Youths with Autism
Applies self-exciting temporal point processes to model clustered aggression onsets in inpatient autistic youth and reports better fit than Poisson baselines.
-
A Uniform Determination of the Bulk Metallicities and Alpha Enrichments of Confirmed Exoplanet Systems with TRES
A uniform spectroscopic catalog of 625 exoplanet hosts shows subsolar-metallicity giant-planet hosts are alpha-enhanced relative to both iron-rich hosts and typical metal-poor field stars.
-
Plato's view on supermassive black hole binaries: Exploring the faint limit of ESA's Plato space mission
Simulations show Plato can recover relativistic photometric signatures of supermassive black hole binaries in bright quasars (G≤18) via Bayesian inference on mock light curves.
-
Mitigating effects of telescope jitter through differentiable forward-modeling
Differentiable optical simulation models telescope jitter blurring and shows that two-dimensional jitter models avoid systematic bias in binary separation measurements for the TOLIMAN exoplanet mission.
-
Fast Bayesian equipment condition monitoring via simulation based inference: applications to heat exchanger health
Amortized neural posterior estimation via simulation-based inference delivers 82x faster inference than MCMC for heat exchanger fouling and leakage diagnosis while maintaining comparable accuracy on synthetic data.
-
Neural posterior estimation for scalable and accurate inverse parameter inference in Li-ion batteries
NPE delivers millisecond-scale parameter inference for Li-ion batteries that matches or exceeds Bayesian calibration accuracy while adding local sensitivity interpretability, though with higher voltage prediction errors.
-
GW250114: testing Hawking's area law and the Kerr nature of black holes
GW250114 data confirm the remnant black hole ringdown frequencies lie within 30% of Kerr predictions and that the final horizon area is larger than the sum of the progenitors' areas to high credibility.
-
Understanding Deterioration Random Effects for Causal Discovery in Infrastructure Management
Bayesian random effects estimation followed by stratified DirectLiNGAM causal discovery on 112 pumps shows 400x larger causal effects from operational features in slower-deteriorating equipment compared to faster-deteriorating ones.