INDEX
Every paper Pith has read and judged, in one searchable index.
-
stat.ME arXiv submitted 2026-07-17One threshold pins exact anomaly detectability in Gaussian data
Marie Tur\v{c}i\v{c}ov\'a +1 · “Asymptotically exact threshold for detecting anomalies in multivariate Gaussian data with application to time series”
2607.15637 -
stat.ML arXiv submitted 2026-07-17Retraining locks onto stable signals even when model feedback is huge
Moritz Hardt · “Retraining Seeks Stable Signals”
2607.15623 -
cs.LG arXiv submitted 2026-07-17Asymmetric nearest-neighbor test detects LLM drift
Sergey Zakharov +2 · “ASK-NN: An Asymmetric Nearest-Neighbor Test that detects Distribution Drifts in Natural Language”
2607.15607 -
cs.LG arXiv submitted 2026-07-17Pooled benchmarks miss temporal failures in synthetic data
Kiwan Kwon +7 · “Do Generative Models Keep Time? A Time-Aware Evaluation of Synthetic Sequential Tabular Data”
2607.15606 -
cs.IT arXiv submitted 2026-07-17Original BIHT hits optimal 1-bit recovery with no normalization
Arya Mazumdar +1 · “On the Role of Normalization in Binary Iterative Hard Thresholding for 1-bit Compressed Sensing”
2607.15530 -
stat.CO arXiv submitted 2026-07-16One extra variable drives MCMC between distant modes
Renny Doig +1 · “Compound Auxiliary Metropolis: Incorporating Auxiliary Variables into Multi-Candidate MCMC”
2607.15499 -
cs.LG arXiv submitted 2026-07-16Diffusion models recover mixture weights the target score can't see
Andrew Dennehy +3 · “Diffusion models recover accurate mixture weights despite score function insensitivity”
2607.15485 -
stat.ML arXiv submitted 2026-07-16Nested audit weights turn noisy labels into valid estimates
Robert Chew +1 · “Design-Based Supervised Learning with Noisy Human Labels”
2607.15455 -
stat.ME arXiv submitted 2026-07-16Five-test panel beats Hosmer–Lemeshow on sparse data
Ebrahim Khaled Ebrahim +1 · “Benchmarking Goodness-of-Fit and Calibration Algorithms for Logistic Regression Classifiers: A Large-Scale Simulation Study under Sparse Data”
2607.16344 -
stat.ME arXiv submitted 2026-07-16Directional correction lets Hosmer-Lemeshow see asymmetric-link misfit
Ebrahim Khaled Ebrahim +1 · “A directional Hosmer-Lemeshow goodness-of-fit test for sparse logistic regression”
2607.15454 -
math.ST arXiv submitted 2026-07-16Mixing teacher with its distilled student beats the teacher
Hien Dang +2 · “Prediction-Only Distillation in Linear and Logistic Regression”
2607.15450 -
stat.AP arXiv submitted 2026-07-16Hourly bed forecasts cut ED boarding by 30–70 percent
Qian Cheng +3 · “Proactive Inpatient Bed Requests for Emergency Department Admissions”
2607.15432 -
eess.SY arXiv submitted 2026-07-16Six Sigma comes to 3D printing: one loop
Hui Yang +7 · “Six-sigma Quality Management of Additive Manufacturing”
2607.15430 -
stat.ME arXiv submitted 2026-07-16Null-quantile cutoff controls false alarms in change point tests
Melissa Lynne Martin +2 · “Sequential Control of False Positives in Online Change Point Detection”
2607.15423 -
cs.LG arXiv submitted 2026-07-16Learn the best block replacement interval with log regret
Aniruddhan Ganesaraman +1 · “Data Driven Block Replacement Scheduling”
2607.15229 -
stat.CO arXiv submitted 2026-07-16O(sqrt K) steps control marginal bias in unadjusted HMC
Yifan Chen +3 · “Delocalization of bias in unadjusted Hamiltonian Monte Carlo and underdamped Langevin”
2607.15208 -
stat.ML arXiv submitted 2026-07-16One decomposition reproduces the field's uncertainty measures
Raghad Alamri +2 · “Subjective Risk Decomposition: A New View for Uncertainty Quantification”
2607.15196 -
stat.ME arXiv submitted 2026-07-16One likelihood couples SEIR epidemics with dynamic contact networks
Md Asaduzzaman · “A Complete-Data Likelihood for Epidemic Processes on Partially Observed Dynamic Networks”
2607.15179 -
stat.ME arXiv submitted 2026-07-16A spike-and-slab prior recovers true signal frequencies
Beniamino Hadj-Amar +2 · “Frequency Selection in Bayesian Spectral Modeling of Time Series Data with Applications to Wearable Device Measurements”
2607.15157 -
stat.ML arXiv submitted 2026-07-16cGAP colors categorical cells by embedding geometry to expose clusters
Chun-houh Chen +7 · “cGAP: Generalized Association Plots with HOMALS-Guided Heatmaps for Visualization of High-Dimensional Categorical Data”
2607.15018 -
stat.ME arXiv submitted 2026-07-16Tiny error-budget shift gives goodness-of-fit tests big power gains
Roman Guchenko · “Priority-preserving augmentation of goodness-of-fit tests by conditional calibration”
2607.15015 -
stat.AP arXiv submitted 2026-07-16One fitted geometric model maps joint flood risk across 10 sites
Kristina Grolmusova +2 · “Flood risk estimation via geometric extremal graphical models”
2607.15000 -
math.ST arXiv submitted 2026-07-16Four coherent measures yield the same experiment
A. Philip Dawid · “Coherent Measures of Discrepancy, Uncertainty and Dependence, with Applications to Bayesian Predictive Experimental Design”
2607.26077 -
math.OC arXiv submitted 2026-07-16Scenario-based optimization errors obey a √N Gaussian-process limit
Aurya Javeed +1 · “Statistical Inference for Scenario-Based Dynamic Optimization under Uncertainty”
2607.14965 -
stat.AP arXiv submitted 2026-07-16Monte Carlo scheme puts error bars on boundary-layer height
Tommaso Locatelli +2 · “Statistical Modelling of Planetary Boundary Layer Height and Its Measurement Uncertainty Using GRUAN Profiles”
2607.14960 -
math.PR arXiv submitted 2026-07-16Self-overlap sets the sharp line for sparse graph alignment
Louis Vassaux · “Graph alignment in sparse inhomogeneous models via self-overlap”
2607.14948 -
stat.ML arXiv submitted 2026-07-16Closed-form mix of true and teacher velocities improves rectified flow
Saptarshi Roy +2 · “Optimal Self-Distillation for Rectified Flow via Linear Probing”
2607.14947 -
math.ST arXiv submitted 2026-07-16Copula regression tests now target the regression function itself
Holger Dette +1 · “Testing for correct model specification in copula regression models”
2607.14930 -
stat.AP arXiv submitted 2026-07-16Budget-constrained life-test design reduces to one equation
Rathin Das +2 · “Optimal Design for Generalized Progressive Hybrid Censored Data via Constrained, Unconstrained, Compound, and Minimax Optimization”
2607.14926 -
math.ST arXiv submitted 2026-07-16Diffusion time to uniformity beats Moran's I on city data
Thomas Weighill +1 · “Measuring Spatial Clustering via Metropolis-Hastings Diffusion Distance”
2607.14880 -
math.OC arXiv submitted 2026-07-16Tamed stochastic-gradient HMC converges at λ^1/4 rate
Zhuoran Wang +1 · “Tamed Stochastic Gradient Hamiltonian Monte Carlo”
2607.14862 -
stat.ME arXiv submitted 2026-07-16No preferential sampling in Tokyo store closures
Akitoshi Kanetaka +1 · “Assing Preferential Sampling in Retail Survival Data: A Bayesian Joint LGCP and Spatial Probit Model for Mini-Supermarket Closure in Tokyo”
2607.14860 -
stat.ME arXiv submitted 2026-07-16A way to say which variables caused a detected change
Dhia-Elhaq Ouerfelli +3 · “Post Hoc Inference for Component Attribution in Multivariate Change-Point Detection”
2607.14814 -
math.ST arXiv submitted 2026-07-16Correlated Gaussian tests can push BH FDR past any fixed multiple of q
Lihua Lei · “How Much Can Gaussian Dependence Inflate the Benjamini-Hochberg Procedure's FDR?”
2607.14812 -
stat.ME arXiv submitted 2026-07-16Neural alignment nets beat linear mixed-frequency forecasts
Yize Wang +2 · “Mixed-Frequency Time Series Forecasting via Depth-Separable Neural Networks”
2607.14771 -
cs.LG arXiv submitted 2026-07-16Grouped Hawkes attention beats six TKG benchmarks
Xiangni Tian +4 · “GAttNHP: Group Attention Neural Hawkes Process for Extrapolation Reasoning in Temporal Knowledge Graphs”
2607.14733 -
cs.LG arXiv submitted 2026-07-16Local SGD provably beats mini-batch when curvatures align
Kumar Kshitij Patel +5 · “What's in a Smoothness Constant? Tighter Rates for Local SGD with Bounded Second-order Heterogeneity”
2607.14731 -
stat.ME arXiv submitted 2026-07-16New test proves when logistic regression is close enough
Vladimir Ostrovski · “Testing equivalence to binary generalized linear models with application to logistic regression”
2607.14724 -
stat.ME arXiv submitted 2026-07-16Exact formula prices Wiener-Hermite mismatch under non-Gaussian inputs
Serhii Zabolotnii · “Exact Computation of Non-Gaussian Mismatch Penalties in Wiener-Hermite Cross-Correlation Identification”
2607.14699 -
math.NA arXiv submitted 2026-07-16Split source and boundary posteriors hit the minimax rate
Emmanuel E. Oguadimma · “Operator-Split Bayesian Learning for Elliptic PDEs with Unequal Interior and Boundary Data”
2607.14680 -
stat.ME arXiv submitted 2026-07-16Multiverse analysis can hide abdication and manufacture doubt
Martin Modr\'ak · “Multiverse analysis, abdication of responsibility and manufacturing of doubt”
2607.14623 -
cs.LG arXiv submitted 2026-07-16Policy overlap unlocks a proven A/B variance cut
Olivier Jeunen · “Accelerating A/B-Tests with Counterfactual Estimation: Reducing Variance through Policy Overlap”
2607.14604 -
cs.LG arXiv submitted 2026-07-16q≤1 decides whether residual nets train stably
Hyemin Gu +3 · “Sharp Stability Threshold and Certification for Designing Stable Residual Architectures”
2607.14576 -
cs.LG arXiv submitted 2026-07-16Recovers elasticity maps from displacement data at 12.5% resolution
Tatthapong Srikitrungruang +1 · “Probabilistic Physics-Informed Neural Networks for Estimating Heterogeneous Elastic Properties from Low-Resolution and Noisy Displacement Data”
2607.14563 -
cs.LG arXiv submitted 2026-07-16RK-Adam's 'adaptivity' is a no-op: it's fixed-step Adam at 3-4x cost
Akhilesh Gogikar · “Adaptive Runge-Kutta Step Control Buys Training Loss, Not Generalization: An Honest Compute-Matched Study of RK-Adam Optimizers”
2607.14516 -
stat.CO arXiv submitted 2026-07-16Custom Gauss rules to 18-digit accuracy from an R call
Paul Kabaila · “Custom-made Gauss quadrature: an introduction for statisticians”
2607.14511 -
stat.ME arXiv submitted 2026-07-16Latent classes track what most tests measure
Nandini Dendukuri +6 · “Improving interpretation of latent class models for diagnostic tests by recognizing their measurands via directed acyclic graphs (DAGs)”
2607.14473 -
math.ST arXiv submitted 2026-07-16Sample covariance error has an exact limit formula
Mihailo Stojnic · “Precise sample covariance spectral norm error -- an RDT view”
2607.14460 -
stat.ME arXiv submitted 2026-07-15Weighted-mean eBH procedures are the only admissible FDR controllers
Liulei Sun +1 · “Admissibility and Complete Classes for False Discovery Rate Control with E-values”
2607.14380 -
cs.LG arXiv submitted 2026-07-15SFT beats ICL once coverage and data quality clear a threshold
Fengzhuo Zhang +2 · “Supervised Fine-Tuning vs. In-Context Learning: An Equilibrium Analysis of LLM Personalization under Congestion”
2607.14371