O'Prior, a compositional synthetic prior with hierarchical SCMs, realism engines, stress modules, and curriculum protocols, improves tabular foundation model accuracy and robustness on real benchmarks when architecture and compute are held fixed.
MIT press Cambridge, MA
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
background 1polarities
support 1representative citing papers
MyoChallenge 2025 introduces standardized table tennis and soccer tasks for musculoskeletal models in the MyoSuite simulation framework to benchmark athletic motor control algorithms.
MetaColloc meta-learns a universal set of neural basis functions offline so that new PDEs can be solved at test time with a single linear solve instead of per-equation neural-network optimization.
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.
A new kernel nonconformity score for multivariate conformal prediction that adapts to residual geometry, provides finite-sample coverage, and achieves convergence rates based on effective kernel rank rather than ambient dimension.
citing papers explorer
-
Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality
O'Prior, a compositional synthetic prior with hierarchical SCMs, realism engines, stress modules, and curriculum protocols, improves tabular foundation model accuracy and robustness on real benchmarks when architecture and compute are held fixed.
-
MyoChallenge 2025: A New Benchmark for Human Athletic Intelligence
MyoChallenge 2025 introduces standardized table tennis and soccer tasks for musculoskeletal models in the MyoSuite simulation framework to benchmark athletic motor control algorithms.
-
MetaColloc: Optimization-Free PDE Solving via Meta-Learned Basis Functions
MetaColloc meta-learns a universal set of neural basis functions offline so that new PDEs can be solved at test time with a single linear solve instead of per-equation neural-network optimization.
-
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.
-
A Kernel Nonconformity Score for Multivariate Conformal Prediction
A new kernel nonconformity score for multivariate conformal prediction that adapts to residual geometry, provides finite-sample coverage, and achieves convergence rates based on effective kernel rank rather than ambient dimension.