Cross-Fitting

Latest papers

9 papers4 papers in the last four weeks, with none the four weeks before. 0.0% of all new papers.

BRIDGE: An Acceptance Gate to Block False Improvements in Model Updates, with a Retrospective Clinical Evaluation
Oct 21, 2026Dwi SunaryonoCross-Fitting Model Selection
Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia
Cross-Fitting for Neural Posterior Estimation
Oct 19, 2026Jeffrey RegierCross-Fitting Neural Posterior Estimation
Department of Statistics, University of Michigan, Ann Arbor, MI 48109
False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents
Sep 30, 2026Meijia Chen, Hao Li, Zheng Lu +12Cross-Fitting Agentic Search
Rutgers University · Independent Researcher · University of California, San Diego +3
Transfer Calibrated Prediction Powered Inference
Sep 28, 2026Aditya T. Vadlamani, Jae Ho Chang, Srinivasan Parthasarathy +1Cross-Fitting Domain Adaptation
The Ohio State University · Yale University
Cross-Fitted Residual Utility for Primary-Preserving Cognitive Decision Correction in Automatic Modulation Classification
Aug 3, 2026Linzhuo Han, Zongyong Cui, Houbiao LiCross-Fitting Automatic Modulation Classification
School of Mathematical Sciences, University of Electronic Science and Technology of China, Chengdu, China · School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China
On Local Population-Risk Certificates
Jun 17, 2026Mingzhi SongCross-Fitting Statistical Learning Theory
Department of Mathematics, The University of Hong Kong, Hong Kong
Improving reproducibility by controlling random seed stability in machine learning based estimation via bagging
Apr 20, 2026Nicholas Williams, Alejandro SchulerCross-Fitting Debiased ML
Division of Biostatistics, University of California, Berkeley, California, U.S.A.
Cross-fitted Proximal Learning for Model-Based Reinforcement Learning
Apr 6, 2026Nishanth Venkatesh, Andreas A. MalikopoulosCross-Fitting Reinforcement Learning
Department of Systems Engineering, Cornell University, Ithaca, NY, USA · Applied Mathematics, Systems Engineering, Mechanical Engineering, Electrical & Computer Engineering, and School of Civil & Environmental Engineering, Cornell University, Ithaca, NY, USA
Functional Natural Policy Gradients
Mar 30, 2026Aurelien Bibaut, Houssam Zenati, Thibaud Rahier +1Cross-Fitting Natural Policy Gradient

Oct 21, 2026J/K move · Enter open · S save

BRIDGE: An Acceptance Gate to Block False Improvements in Model Updates, with a Retrospective Clinical Evaluation

Dwi Sunaryono

Department of Informatics, Faculty of Intelligent Electrical and Informatics Technology, Institut Teknologi Sepuluh Nopember, Surabaya 60111, Indonesia

Cross-Fitting Model Selection Selective Inference

Open paper AskPDF arXiv

Models are updated repeatedly, and an update that looks better on validation data but not on new data is a false improvement that can misguide clinical decisions. The naive rule accepts every positive validation gain and is optimistic when the same data choose and judge the update. BRIDGE is a model-agnostic acceptance gate that decides each update from predictions and the available labels, with fixed settings. Each fold of the data judges the update chosen on the other folds (cross-fitting) to reduce selection bias; acceptance requires the gain's lower bootstrap bound to exceed a margin. The four settings are (A) 11 new datasets, (B) synthetic tasks, (C) 12 retrospective clinical tasks, and (D) 39 supporting datasets. The false-improvement rate falls from 19.1% (naive) to 1.1% (BRIDGE) in (A), 20.7% to 1.7% (B), 20.0% to 0% (C), and 12.5% to 0.5% (D); every 95% confidence interval of the reduction over datasets or tasks excludes zero (not over the clinical sources). An equal-size random draw from the naive acceptances contains more false improvements in (A)-(E) (post hoc, p<0.001). On the OpenML collection (E) of 14 new datasets, the false-improvement rate falls from 25.6% to 1.0%. BRIDGE removes 97% to 100% of the false improvements that the naive rule accepts, and its mean gain per deployed update is 1.8 to 3.1 times the naive rule's. The 95% upper bound on the false-improvement rate is below 5.5%: formal in (B) for independent tasks (3.56%), and conditional on the real data tested (0.33% to 5.40%).