Four types of LLM reliance (Strategic 34.3%, Instrumental 30.9%, Dialogic 30.4%, Dependent 4.5%) were identified among undergraduates, with AI literacy predicting type and value/cost beliefs predicting intensity.
hub
Tree-Structured Classification via Generalized Discriminant Analysis
13 Pith papers cite this work. Polarity classification is still indexing.
13
Pith papers citing it
hub tools
verdicts
UNVERDICTED 13representative citing papers
Introduces zero-inflated Gaussian distributions for EDAs to jointly optimize sparsity patterns and active parameter values without bi-level schemes or custom operators.
TiRex-2 is a recurrent xLSTM time series foundation model for multivariate forecasting with future covariates and constant-cost streaming that reports SOTA zero-shot results on GIFT-Eval and fev-bench.
Kolmogorov n-width theory plus PRESS statistics yield closed-form optimal spline resolution; KORE estimates bias/noise scales from two pilots and matches CV performance with far fewer fits.
Predictively consistent priors let complex Bayesian models match or beat the out-of-sample performance of selected simpler models across linear, logistic, and nonlinear examples without explicit selection.
PliableBVS is a new Bayesian hierarchical spike-and-slab model for simultaneous selection of high-dimensional main effects and interactions under an asymmetric weak hierarchical constraint, shown to outperform pliable lasso in simulations.
A Bayesian search calibrated on 3,493 meteorite analyses finds strong evidence for a siderophile-enriched processed template in only a small fraction of 697 polluted white dwarf records, with posterior medians of 0.011 and 0.041.
Proves well-posedness and unique invariant measure (delta at 0 plus gamma density) for sticky CIR, derives Green's function for exact sampling, and analyzes MH and ULA samplers with explicit bias for the potential case via Girsanov.
Affordance-driven Croissant Charts improve performance on probability comparison tasks with normal distributions, validated in a preregistered study with 808 participants.
Unifying framework for CTree, MOB and GUIDE shows model scores without dichotomization yield higher power for covariate selection than residuals or dichotomized scores in many scenarios.
CMS excludes narrow-width singlet vector-like B' quarks with masses 0.8-1.23 TeV in the tW channel using neural-network tagging and data-driven background modeling.
Proves oracle Bernstein-von Mises theorem for fractional posterior under supportwise likelihood assumptions in sparse GLMs with spike-and-slab priors.
FIRE-2 simulations show Milky Way analogs form in two phases whose transition correlates with the host halo's shift from fast to slow accretion, driven by two distinct star-formation modes.
citing papers explorer
-
Four Types of LLM Reliance and Their Predictors Among Undergraduate Writers: A Mixed-Methods Study at a Minority-Serving R1 University
Four types of LLM reliance (Strategic 34.3%, Instrumental 30.9%, Dialogic 30.4%, Dependent 4.5%) were identified among undergraduates, with AI literacy predicting type and value/cost beliefs predicting intensity.
-
Zero-Inflated Gaussian Distributions Enable Parameter-Space Sparsity in Estimation-of-Distribution Algorithms
Introduces zero-inflated Gaussian distributions for EDAs to jointly optimize sparsity patterns and active parameter values without bi-level schemes or custom operators.
-
TiRex-2: Generalizing TiRex to Multivariate Data and Streaming
TiRex-2 is a recurrent xLSTM time series foundation model for multivariate forecasting with future covariates and constant-cost streaming that reports SOTA zero-shot results on GIFT-Eval and fev-bench.
-
Solve for the Hyperparameter, Skip the Search: Kolmogorov-Optimal Scaling Laws for Spline Regression
Kolmogorov n-width theory plus PRESS statistics yield closed-form optimal spline resolution; KORE estimates bias/noise scales from two pilots and matches CV performance with far fewer fits.
-
To select or not to select: predictively consistent priors instead of model selection
Predictively consistent priors let complex Bayesian models match or beat the out-of-sample performance of selected simpler models across linear, logistic, and nonlinear examples without explicit selection.
-
PliableBVS: A flexible Bayesian variable selection method for modeling interactions with mandatory modifying variables
PliableBVS is a new Bayesian hierarchical spike-and-slab model for simultaneous selection of high-dimensional main effects and interactions under an asymmetric weak hierarchical constraint, shown to outperform pliable lasso in simulations.
-
A Calibrated Bayesian Search for Potential Chemical Technosignatures in Polluted White Dwarf
A Bayesian search calibrated on 3,493 meteorite analyses finds strong evidence for a siderophile-enriched processed template in only a small fraction of 697 polluted white dwarf records, with posterior medians of 0.011 and 0.041.
-
Sticky CIR process with potential: invariant measure and exact sampling
Proves well-posedness and unique invariant measure (delta at 0 plus gamma density) for sticky CIR, derives Green's function for exact sampling, and analyzes MH and ULA samplers with explicit bias for the potential case via Girsanov.
-
Croissant Charts: Modulating the Performance of Normal Distribution Visualizations with Affordances
Affordance-driven Croissant Charts improve performance on probability comparison tasks with normal distributions, validated in a preregistered study with 808 participants.
-
The Power of Unbiased Recursive Partitioning: A Unifying View of CTree, MOB, and GUIDE
Unifying framework for CTree, MOB and GUIDE shows model scores without dichotomization yield higher power for covariate selection than residuals or dichotomized scores in many scenarios.
-
Search for single production of a vector-like B' quark decaying to a top quark and a W boson in the single-lepton final state in proton-proton collisions at $\sqrt{s}$ = 13 TeV
CMS excludes narrow-width singlet vector-like B' quarks with masses 0.8-1.23 TeV in the tW channel using neural-network tagging and data-driven background modeling.
-
Bernstein-von Mises Theorem for Sparse Generalized Linear Model
Proves oracle Bernstein-von Mises theorem for fractional posterior under supportwise likelihood assumptions in sparse GLMs with spike-and-slab priors.
-
Two-phase formation of galaxies: the coevolution between galaxies and dark matter halos
FIRE-2 simulations show Milky Way analogs form in two phases whose transition correlates with the host halo's shift from fast to slow accretion, driven by two distinct star-formation modes.