Clay weekly context brief for the Statistics category (ISO week 2026-W30).

Clay tracks publications from the Statistics feed list. Below are recent items from this category, each with its source and a short description of what the publication covers when one is available in the source feed.

Recent publications:
1. Latency-Response Theory Model: Evaluating Large Language Models via Response Accuracy and Chain-of-Thought Length
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2512.07019
   The proliferation of Large Language Models (LLMs) necessitates valid evaluation methods to provide guidance for both downstream applications and actionable future improvements.
2. Bayesian Variable Selection in Generalized Linear Models
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2606.24357
   Covariate selection in Generalized Linear Models (GLMs) is a fundamental problem in statistics, as including irrelevant predictors might lead to overfitting and poor interpretability, while omitting relevant ones might result in biased estimates.
3. Manifold Dimension Estimation via Local Graph Structure
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2510.15141
   Most existing manifold dimension estimators rely on the assumption that the underlying manifold is locally flat within the neighborhoods under consideration.
4. Modeling group heterogeneity in spatio-temporal data via physics-informed semiparametric regression
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2511.13203
   In this work we propose a novel approach for modeling spatio-temporal data characterized by group structures.
5. Flood risk estimation via geometric extremal graphical models
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.15000
   We exploit the new framework of multivariate geometric extreme value theory for the statistical analysis of river flow extremes at multiple locations on a river network.
6. Analysis of Semi-Supervised Learning on Hypergraphs
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2510.25354
   Hypergraphs provide a natural framework for modeling multiway interactions.
7. A model-based restricted shapley value to measure the players' contribution to shot actions in football
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2603.11016
   This paper proposes a novel framework to assess individual player contributions in football, explicitly accounting for the cooperative nature of shot-ending offensive actions.
8. Learning What to Learn: Experimental Design when Combining Experimental with Observational Evidence
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2510.23434
   Experiments deliver credible treatment-effect estimates but, because they are costly, are often restricted to specific sites, small populations, or particular mechanisms.
9. The Internet of Things for Smart Manufacturing: A Review
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.16172
   The modern manufacturing industry is investing in new technologies such as the Internet of Things (IoT), big data analytics, cloud computing and cybersecurity to cope with system complexity, increase information visibility, improve production performance, and gain competitive advantages in the global market.
10. Detection of collective and point anomalies in the presence of trend and seasonality
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2508.21128
   Detecting anomalies in time series data is a challenging task with broad relevance in many applications.
11. Scaling Hawkes Processes
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.16081
   Hawkes processes (HP) are a large class of stochastic point process models scientists have used to analyze contagion phenomena ranging from earthquakes, infectious diseases and biological neurons to financial trading activity, memes on social media and gun violence.
12. Consistent Bayesian Spatial Domain Partitioning Using Predictive Spanning Tree Methods
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2508.08324
   Bayesian model-based spatial clustering methods are widely used for their flexibility in estimating latent clusters with an unknown number of clusters while accounting for spatial proximity.
13. Deep and Probabilistic Models for Gene Regulatory Network Inference
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.16053
   Gene regulatory networks (GRNs) link transcription factor (TF) proteins to their target genes, yet reconstructing these networks from genome-wide data remains challenging under practical and methodological constraints.
14. Minimax and Bayes Optimal Best-Arm Identification
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2506.24007
   This study investigates minimax and Bayes optimal strategies for fixed-budget best-arm identification.
15. Design-Based Supervised Learning with Noisy Human Labels
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.15455
   Researchers increasingly use automated classifiers to label unstructured data for statistical analysis.
16. Testing Separability of High-Dimensional Covariance Matrices
   Source: stat.TH (Statistics Theory)
   Link: https://arxiv.org/abs/2506.17463
   Due to their parsimony, separable covariance models have been popular in modeling matrix-variate data.
17. A directional Hosmer-Lemeshow goodness-of-fit test for sparse logistic regression
   Source: stat.AP (Applications)
   Link: https://arxiv.org/abs/2607.15454
   Goodness-of-fit assessment for the binary logistic regression model is difficult when covariates are continuous: the data are effectively sparse, the classical Pearson and deviance tests fail, and practitioners rely on partition-based tests, such as the Hosmer-Lemeshow test, that group observations before comparing observed and expected counts.

Sources in this brief: stat.AP (Applications); stat.TH (Statistics Theory).
Selected 17 of 26 available items for this weekly brief.