MulTaBench is a new collection of 40 image-tabular and text-tabular datasets designed to test target-aware representation tuning in multimodal tabular models.
mega hub Mixed citations
Random forests
Mixed citation behavior. Most common role is background (57%).
36
Pith papers citing it
110k
external citations · Crossref
Background
57% of classified citations
hub tools
citation-role summary
background 4
method 2
baseline 1
citation-polarity summary
authors
mega hub controls
Recognition alignment
counterfactual ablation
co-cited works
representative citing papers
LG-CoTrain, an LLM-guided co-training method, outperforms classical semi-supervised baselines for crisis tweet classification in low-resource settings with 5-25 labeled examples per class.
An intrinsic effective sample size for manifold MCMC is defined via kernel discrepancy as the number of independent draws yielding equivalent expected squared discrepancy to the target.
EstGraph benchmark evaluates LLMs on estimating properties of very large graphs from random-walk samples that fit in context limits.
The profile maximum likelihood estimator for the location in anisotropic hyperbolic wrapped normal models is strongly consistent, asymptotically normal, and attains the Hájek-Le Cam minimax lower bound under squared geodesic loss.
SynQL synthesizes diverse, execution-ready SQL workloads by deterministically traversing foreign-key graphs to populate ASTs, yielding high topological entropy and cost-model training data with R² ≥ 0.79 on held-out sets.
RCT couples an LLM and Random Forest via RL feedback so each augments the other's features and rewards, producing consistent gains on three medical datasets.
Semantic segmentation decomposes monitoring features into canonical and residual components that concentrate fault-predictive information while preserving operational meaning in predictive maintenance.
Machine translation preserves embedding similarity structure for ten languages but distorts it for four in the Manifesto Corpus, via a new non-inferiority testing framework.
Vesselpose predicts voxel-wise direction vectors to extend the TEASAR algorithm for topologically accurate vascular graph reconstruction from 3D images.
RCProb uses Dirichlet-smoothed class priors and Beta-smoothed condition likelihoods in a Naive Bayes formulation to extract rules from tree ensembles approximately 22 times faster than RuleCOSI+ while maintaining competitive accuracy and producing more compact rule sets on 33 benchmark datasets.
StarCLR pretrains on TESS light curves via contrastive learning on overlapping subsequences and improves variable star classification F1 scores over scratch-trained models when fine-tuned on TESS, ZTF, and Gaia.
Random forests on string similarity features outperform LLMs for German dialect lexicon induction and boost dialect information retrieval by up to 50% in recall.
Foundation models outperform dataset-specific machine learning in energy time series forecasting across 54 datasets in 9 categories.
ReSS uses decision-tree scaffolds to fine-tune LLMs for faithful tabular reasoning, reporting up to 10% gains over baselines on medical and financial data.
A criterion of |Δg| > 0.4 mag and |Δ(g-r)| > 0.2 mag detects photometric CL-AGN transitions in 9.6% of known hosts with 1.6% false positive rate from simulations.
Entity recognition models detect ads in RAG responses effectively and stay robust when advertisers switch styles, while lightweight models like random forests and SVMs become brittle under the same changes.
Neural network classification with CRPS optimization produces calibrated photometric redshift PDFs for DESI Legacy and Pan-STARRS data, achieving σ_NMAD of 0.0153 on LSDR10 and outperforming regression methods.
The synthetic prior for tabular foundation models covers only a narrow part of real table distributions, but this mismatch does not degrade model generalization.
Physics-guided data augmentation combined with neural networks enables accurate indentation size effect correction in steels from small sets of shallow nanoindentation measurements, outperforming Nix-Gao in the shallow regime.
A knowledge-data dual paradigm using geomorphic priors and a tabular foundation model achieves baseline-level landslide susceptibility prediction accuracy with only 30% of typical data in tested regions.
A novel algorithm learns sets of optimal quantile regression trees to predict full conditional distributions interpretably and efficiently.
An ensemble ML framework achieves 90.7% morphology classification accuracy and R² values of 0.77–0.92 for key parameters on held-out test data, with external validation against OGLE and Kepler catalogs.
A transit search on TESS Cycle 1 full-frame images produced 10,091 new planet candidates down to T=16 mag, more than doubling the known TESS total, with one hot Jupiter confirmed by radial velocity.
citing papers explorer
-
MulTaBench: Benchmarking Multimodal Tabular Learning with Text and Image
MulTaBench is a new collection of 40 image-tabular and text-tabular datasets designed to test target-aware representation tuning in multimodal tabular models.
-
LLM-guided Semi-Supervised Approaches for Social Media Crisis Data Classification
LG-CoTrain, an LLM-guided co-training method, outperforms classical semi-supervised baselines for crisis tweet classification in low-resource settings with 5-25 labeled examples per class.
-
Intrinsic effective sample size for manifold-valued Markov chain Monte Carlo via kernel discrepancy
An intrinsic effective sample size for manifold MCMC is defined via kernel discrepancy as the number of independent draws yielding equivalent expected squared discrepancy to the target.
-
Evaluating LLMs on Large-Scale Graph Property Estimation via Random Walks
EstGraph benchmark evaluates LLMs on estimating properties of very large graphs from random-walk samples that fit in context limits.
-
Profile Likelihood Inference for Anisotropic Hyperbolic Wrapped Normal Models on Hyperbolic Space
The profile maximum likelihood estimator for the location in anisotropic hyperbolic wrapped normal models is strongly consistent, asymptotically normal, and attains the Hájek-Le Cam minimax lower bound under squared geodesic loss.
-
SynQL: A Controllable and Scalable Rule-Based Framework for SQL Workload Synthesis for Performance Benchmarking
SynQL synthesizes diverse, execution-ready SQL workloads by deterministically traversing foreign-key graphs to populate ASTs, yielding high topological entropy and cost-model training data with R² ≥ 0.79 on held-out sets.
-
Reciprocal Co-Training (RCT): Coupling Gradient-Based and Non-Differentiable Models via Reinforcement Learning
RCT couples an LLM and Random Forest via RL feedback so each augments the other's features and rewards, producing consistent gains on three medical datasets.
-
Semantic Feature Segmentation for Interpretable Predictive Maintenance in Complex Systems
Semantic segmentation decomposes monitoring features into canonical and residual components that concentrate fault-predictive information while preserving operational meaning in predictive maintenance.
-
Is Textual Similarity Invariant under Machine Translation? Evidence Based on the Political Manifesto Corpus
Machine translation preserves embedding similarity structure for ten languages but distorts it for four in the Manifesto Corpus, via a new non-inferiority testing framework.
-
Vesselpose: Vessel Graph Reconstruction from Learned Voxel-wise Direction Vectors in 3D Vascular Images
Vesselpose predicts voxel-wise direction vectors to extend the TEASAR algorithm for topologically accurate vascular graph reconstruction from 3D images.
-
RCProb: Probabilistic Rule Extraction for Efficient Simplification of Tree Ensembles
RCProb uses Dirichlet-smoothed class priors and Beta-smoothed condition likelihoods in a Naive Bayes formulation to extract rules from tree ensembles approximately 22 times faster than RuleCOSI+ while maintaining competitive accuracy and producing more compact rule sets on 33 benchmark datasets.
-
StarCLR: Contrastive Learning Representation for Astronomical Light Curves
StarCLR pretrains on TESS light curves via contrastive learning on overlapping subsequences and improves variable star classification F1 scores over scratch-trained models when fine-tuned on TESS, ZTF, and Gaia.
-
Resource-Lean Lexicon Induction for German Dialects
Random forests on string similarity features outperform LLMs for German dialect lexicon induction and boost dialect information retrieval by up to 50% in recall.
-
FETS Benchmark: Foundation Models Outperform Dataset-specific Machine Learning in Energy Time Series Forecasting
Foundation models outperform dataset-specific machine learning in energy time series forecasting across 54 datasets in 9 categories.
-
ReSS: Learning Reasoning Models for Tabular Data Prediction via Symbolic Scaffold
ReSS uses decision-tree scaffolds to fine-tune LLMs for faithful tabular reasoning, reporting up to 10% gains over baselines on medical and financial data.
-
Identifying Changing-Look AGN Transitions in Light Curve Data with the Zwicky Transient Facility
A criterion of |Δg| > 0.4 mag and |Δ(g-r)| > 0.2 mag detects photometric CL-AGN transitions in 9.6% of known hosts with 1.6% false positive rate from simulations.
-
Detecting RAG Advertisements Across Advertising Styles
Entity recognition models detect ads in RAG responses effectively and stay robust when advertisers switch styles, while lightweight models like random forests and SVMs become brittle under the same changes.
-
Photometric Redshift PDFs via Neural Network Classification for DESI Legacy Imaging Surveys and Pan-STARRS
Neural network classification with CRPS optimization produces calibrated photometric redshift PDFs for DESI Legacy and Pan-STARRS data, achieving σ_NMAD of 0.0153 on LSDR10 and outperforming regression methods.
-
Mind the Gap? A Distributional Comparison of Real and Synthetic Priors for Tabular Foundation Models
The synthetic prior for tabular foundation models covers only a narrow part of real table distributions, but this mismatch does not degrade model generalization.
-
Data-Efficient Indentation Size Effect Correction in Steels Using Machine Learning and Physics-Guided Augmentation
Physics-guided data augmentation combined with neural networks enables accurate indentation size effect correction in steels from small sets of shallow nanoindentation measurements, outperforming Nix-Gao in the shallow regime.
-
Knowledge-Data Dually Driven Paradigm for Accurate Landslide Susceptibility Prediction under Data-Scarce Conditions Using Geomorphic Priors and Tabular Foundation Model
A knowledge-data dual paradigm using geomorphic priors and a tabular foundation model achieves baseline-level landslide susceptibility prediction accuracy with only 30% of typical data in tested regions.
-
Interpretable Quantile Regression by Optimal Decision Trees
A novel algorithm learns sets of optimal quantile regression trees to predict full conditional distributions interpretably and efficiently.
-
Is the `Known' Enough? An Integrated Machine Learning Framework for Eclipsing Binary Classification and Parameter Estimation Based on Well-Characterized Systems
An ensemble ML framework achieves 90.7% morphology classification accuracy and R² values of 0.77–0.92 for key parameters on held-out test data, with external validation against OGLE and Kepler catalogs.
-
The T16 Planet Hunt: 10,000 New Planet Candidates from TESS Cycle 1 and the Confirmation of a Hot Jupiter Around TIC 183374187
A transit search on TESS Cycle 1 full-frame images produced 10,091 new planet candidates down to T=16 mag, more than doubling the known TESS total, with one hot Jupiter confirmed by radial velocity.
-
Financial Dynamics and Interconnected Risk of Liquid Restaking
Renzo liquid restaking revenue is primarily predicted by EigenLayer value locked, token yield, and multi-blockchain expansion, with current bridge risks not imposing systemic threats to the restaking ecosystem.
-
On Improving Graph Neural Networks for QSAR by Pre-training on Extended-Connectivity Fingerprints
Pre-training GNNs on ECFP prediction produces statistically significant QSAR gains on five of six Biogen benchmarks with OOD splits, but underperforms on heterogeneous datasets and complex endpoints like binding affinity.
-
Generating Synthetic Malware Samples Using Generative AI
Opcode-sequence generative models produce synthetic malware data that raises minor-class classification accuracy by up to 60% and overall detection to 96%.
-
Predicting Redshift in Seyfert Galaxies Using Machine Learning
Random Forest regression on combined optical plus mid-infrared colors yields NMAD of 0.0188, R-squared of 0.9561, and 0.294 percent outliers for photometric redshifts in 23,797 Seyfert II galaxies selected from SDSS and WISE.
-
Implicit neural representations as a coordinate-based framework for continuous environmental field reconstruction from sparse ecological observations
Implicit neural representations enable stable, resolution-independent reconstruction of continuous environmental fields from sparse and irregular ecological data.
-
Impact of Validation Strategy on Machine Learning Performance in EEG-Based Alcoholism Classification
Nested cross-validation reveals optimistic bias in standard validation for EEG alcoholism classification, with AdaBoost reaching 78.3% accuracy and most model differences not statistically significant per McNemar's test.
-
Comparative Evaluation of Machine Learning Models for Predicting Donor Kidney Discard
On 4080 German deceased donors, an ensemble ML model reached MCC 0.76 for kidney discard prediction, with standardized preprocessing and feature selection proving more important than the specific algorithm chosen.
-
A Proof-of-Concept Simulation-Driven Digital Twin Framework for Decision-Aware Diabetes Modeling
A simulation-driven digital twin framework is shown to generate interpretable diabetes trajectories for decision-aware analysis by combining benchmark data with controlled synthetic scenarios.
-
An Explainable Unsupervised-to-Supervised Machine Learning Framework for Dietary Pattern Discovery Using UK National Dietary Survey Data
An unsupervised-to-supervised ML pipeline on UK NDNS data discovers four dietary patterns, reproduces them with macro-F1 0.963 using a surrogate classifier, and interprets them via SHAP for potential clinical use.
-
STRIKE: Additive Feature-Group-Aware Stacking Framework for Credit Default Prediction
STRIKE improves credit default prediction AUC-ROC by training independent models on feature groups and aggregating their outputs via a meta-learner, outperforming tree baselines and conventional stacking on three real datasets.
-
fastml: Guarded Resampling Workflows for Safer Automated Machine Learning in R
fastml is an R package that enforces leakage-free preprocessing through guarded resampling and provides a unified interface for safer automated ML including survival analysis.
- Model Internal Sleuthing: Finding Lexical Identity and Inflectional Features in Modern Language Models