{"total":20,"items":[{"citing_arxiv_id":"2605.21241","ref_index":73,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Divide and Contrast: Learning Robust Temporal Features without Augmentation","primary_cat":"cs.LG","submitted_at":"2026-05-20T14:31:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Di-COT is an unsupervised contrastive method that stochastically partitions time-series windows into overlapping sub-blocks to learn representations without augmentation, reporting SOTA results on classification and transfer tasks across multiple benchmarks while cutting training time.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.19618","ref_index":4,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"A Family of Divergence Measures for Evaluating the Reconstruction Quality of Explainable Ensemble Trees","primary_cat":"cs.LG","submitted_at":"2026-05-19T09:56:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Introduces nLoI and four complementary divergence measures with within/between-node decomposition and unified permutation testing to evaluate surrogate reconstruction quality for Explainable Ensemble Trees.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.19366","ref_index":157,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Accurate, Efficient, and Explainable Deep Learning Approaches for Environmental Science Problems","primary_cat":"cs.LG","submitted_at":"2026-05-19T04:58:10+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The work introduces WaLeF/FIDLAr for flood forecasting, CoDiCast for probabilistic weather, and Hypercube-RAG for explainable environmental QA, claiming superior accuracy, efficiency, and interpretability over baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.19343","ref_index":8,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"What Makes a Representation Good for Single-Cell Perturbation Prediction?","primary_cat":"cs.LG","submitted_at":"2026-05-19T04:30:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"PerturbedVAE disentangles perturbation-specific signals from invariant gene expression structure to recover causal representations and improve out-of-distribution prediction in single-cell perturbation modeling.","context_count":1,"top_context_role":"background","top_context_polarity":"unclear","context_text":"ablesxandualone, even in the simple nonlinear ICA set- ting (Hyvärinen & Pajunen, 1999; Hyvarinen & Morioka, 2016; 2017). To enable the theoretical analysis that fol- lows, we therefore impose additional structure and param- eterize the proposed causal generative model as follows. zι :=λ ιι zι +n ι,n ι ∼ N µι,diagβ ι \u0001 ,(6) zν :=λ νι(u)z ι +λ νν (u)z ν +n ν,(7) nν ∼ N µν(u),diagβ ν(u) \u0001 ,(8) x:= g(z),(9) where, •n ι ∈R dι andn ν ∈R dν are latent noise vari- ables, sampled from Gaussian, i.e.,N µι,diagβ ι \u0001 andN µν(u),diagβ ν(u) \u0001 , respectively. • The matricesλ ιι,λ νν (u), andλ νι(u)are weight ma- trices and are assumed to be strictly lower triangular to satisfy the DAG constraint.1 Here we assume linear causal relationships among the latent variables, pri-"},{"citing_arxiv_id":"2605.16885","ref_index":33,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"A Workflow for Evaluating Regional Treatment Effect Heterogeneity in Multi-Regional Clinical Trials","primary_cat":"stat.AP","submitted_at":"2026-05-16T08:58:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"The paper introduces a question-driven framework and set of statistical methods for exploratory assessment of regional treatment effect heterogeneity in multi-regional clinical trials, evaluated via simulations under no-heterogeneity and modifier-driven scenarios.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.13131","ref_index":154,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"ERPPO: Entropy Regularization-based Proximal Policy Optimization","primary_cat":"cs.LG","submitted_at":"2026-05-13T08:01:20+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"ERPPO adds a DSA-based ambiguity estimator to MAPPO and switches between L1 and L2 entropy regularization to improve exploration and stability in non-stationary multi-dimensional observations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11806","ref_index":214,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Adaptive Kernel Ridge Regression with Linear Structure: Sharp Oracle Inequalities and Minimax Optimality","primary_cat":"math.ST","submitted_at":"2026-05-12T09:00:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"An augmented kernel ridge regression estimator separates linear and nonlinear components to achieve sharp oracle inequalities and minimax optimal prediction risk under general kernels.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11373","ref_index":18,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Causal Algorithmic Recourse: Foundations and Methods","primary_cat":"cs.AI","submitted_at":"2026-05-12T00:55:19+00:00","verdict":"CONDITIONAL","verdict_confidence":"MODERATE","novelty_score":8.0,"formal_verification":"none","one_line_summary":"A causal process model for algorithmic recourse introduces post-recourse stability conditions and copula-based methods to infer intervention effects from observational or paired data, with a distribution-free fallback when the model is rejected.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Generally, counterfactual queries that belong to the so-called third layer of Pearl's Causal Hierarchy2 (PCH, for short) are more difficult to infer compared to interventional queries from the second layer of the PCH (Bareinboim et al., 2022). Therefore, it may be tempting to search for a different formulation of causal recourse, based purely on interventional reasoning, such as computing: P(by|do(R=r), V\\R=v\\r).(18) 1. In this sense, we can see how the setting of algorithmic recourse may differ from the standard literature in causal effect identification, in which none of the mechanismsFof an SCM are known. 2. We remind the reader that the Pearl's Causal Hierarchy consists of the (1)observational, (2)interventional, and (3)counterfactuallayers (Bareinboim et al."},{"citing_arxiv_id":"2605.10315","ref_index":13,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Active Tabular Augmentation via Policy-Guided Diffusion Inpainting","primary_cat":"cs.LG","submitted_at":"2026-05-11T10:17:26+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"TAP couples a learner-conditioned policy with diffusion inpainting to generate and selectively inject high-utility tabular augmentations, yielding up to 15.6 pp accuracy gains and 32% RMSE reduction on seven datasets under severe scarcity.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"peated forward passes on a focused query set. Focused plug-in loss and plug-in utility.Let Qhard(Dt)⊆ Qreal be a subset of informative queries selected using the current evaluator. For classification, we select high-entropy queries. For regression, we select high-uncertainty queries. We define the focused plug-in loss bLψ(Dt) := 1 |Qhard(Dt)| X (x,y)∈Qhard(Dt) ℓ(fψ(Dt)(x), y). (13) Here ψ is an online evaluation procedure andψ(Dt) denotes the evaluator conditioned on the current training set Dt. We use TabPFN (Hollmann et al., 2025) as the default evaluator because it supports fast in-context conditioning and repeated forward passes. bLψ is used only as a ranking signal for can- didate pools during policy learning, and the reported gains"},{"citing_arxiv_id":"2605.09300","ref_index":54,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Causal Stability Selection","primary_cat":"stat.ME","submitted_at":"2026-05-10T03:52:29+00:00","verdict":"CONDITIONAL","verdict_confidence":"MODERATE","novelty_score":7.0,"formal_verification":"none","one_line_summary":"Causal stability selection produces effect-modifier sets with explicit non-asymptotic false-positive bounds by combining cross-fitted CATE estimation and integrated path stability selection.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07285","ref_index":63,"ref_count":2,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Transporting treatment effects by calibrating large-scale observational outcomes","primary_cat":"stat.ME","submitted_at":"2026-05-08T05:48:07+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"A calibration procedure yields a weighted transported average treatment effect with asymptotically valid and efficient inference when experimental data grows slower than observational data, even without positivity or correct OLS specification.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.07096","ref_index":21,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Query-efficient model evaluation using cached responses","primary_cat":"cs.LG","submitted_at":"2026-05-08T01:24:06+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DKPS-based methods predict new model benchmark scores using cached responses, matching baseline mean absolute error with substantially fewer queries and an offline query selection approach.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05984","ref_index":55,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Separable Effects in Four-Arm and Two-Arm Designs","primary_cat":"stat.ME","submitted_at":"2026-05-07T10:32:18+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A methodological framework for separable effects analysis that distinguishes four-arm and two-arm designs, with EIF-based estimation and falsification tests.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.04838","ref_index":9,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data","primary_cat":"stat.ME","submitted_at":"2026-05-06T12:34:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"PAIR-CI restores calibration to conditional independence testing under missing data by using paired permutations that force imputation error to cancel in the loss difference, together with a consistent variance estimator that unifies cross-validation and imputation uncertainty.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.04643","ref_index":183,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Graph-Augmented LLMs for Swiss MP Ideology Prediction","primary_cat":"cs.CL","submitted_at":"2026-05-06T08:39:35+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Graph-augmented LLMs using a political knowledge graph improve ideology prediction accuracy for Swiss MPs by incorporating relational data beyond text alone.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.03141","ref_index":46,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"In-Sample Evaluation of Subgroups Identified by Generic Machine Learning","primary_cat":"stat.ME","submitted_at":"2026-05-04T20:30:25+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A conditional adaptive perturbation approach enables valid in-sample inference for machine learning-identified subgroups with nonregular boundaries via triple robustness.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20288","ref_index":161,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Generative Augmentation of Imbalanced Flight Records for Flight Diversion Prediction: A Multi-objective Optimisation Framework","primary_cat":"cs.LG","submitted_at":"2026-04-22T07:35:11+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Hyperparameter-optimized generative models augment scarce flight diversion records and substantially improve prediction accuracy over real data alone.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.19722","ref_index":3,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Adaptive MSD-Splitting: Enhancing C4.5 and Random Forests for Skewed Continuous Attributes","primary_cat":"cs.LG","submitted_at":"2026-04-21T17:48:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Adaptive MSD-Splitting improves C4.5 and Random Forest performance on skewed data by adjusting standard deviation multipliers for discretization while retaining linear time complexity.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.19477","ref_index":7,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Deep Supervised Contrastive Learning of Pitch Contours for Robust Pitch Accent Classification in Seoul Korean","primary_cat":"cs.SD","submitted_at":"2026-04-21T13:59:37+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Dual-Glob applies supervised contrastive learning to classify fine-grained pitch accent patterns from F0 contours in Seoul Korean, achieving 77.75% accuracy and 51.54% F1 on a new dataset of 10,093 manually annotated accentual phrases.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.18966","ref_index":220,"ref_count":1,"confidence":0.55,"is_internal_anchor":false,"paper_title":"Self-Improving Tabular Language Models via Iterative Reward-Guided Post-Training","primary_cat":"cs.LG","submitted_at":"2026-04-21T01:29:52+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"TabGRAA applies group-relative advantage alignment in an iterative reward-guided post-training loop to improve tabular language model generators on fidelity, utility, and privacy trade-offs across five benchmarks.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}