A new corpus of 108 mixed string-numeric tables shows that advanced tabular learners with basic string embeddings perform well on most real-world data, while large LLM encoders help on free-text heavy tables.
super hub Mixed citations
Gradient-based learning applied to document recognition
Mixed citation behavior. Most common role is background (47%).
82
Pith papers citing it
44.7k
external citations · Crossref
Background
47% of classified citations
hub tools
citation-role summary
background 8
dataset 4
method 3
citation-polarity summary
authors
co-cited works
representative citing papers
SOCP uses self-organizing maps for unsupervised group discovery to enable local calibration in conformal prediction, reducing regional coverage gaps on benchmarks with small set-size increases while preserving validity guarantees.
Dolph2Vec is the first species-specific self-supervised model for dolphin vocalizations, trained on longitudinal recordings from five dolphins, that outperforms general baselines on signature whistle classification and detection while producing embeddings aligned with known whistle categories.
A U-Net surrogate with multigroup attention pooling is trained on OpenMC sensitivity data and combined with gradient optimization to generate grid-based critical experiment geometries that achieve c_k values up to 0.97757 for HALEU fuel validation.
DELOS applies contrastive learning to phase-folded light curves to detect shallow intermediate-to-long period transits, reporting 15.5% and 11.25% gains in combined precision-recall over BLS and TLS in low-SNR tests plus 3-80x speedups.
LineFit delivers more stable line-core intensity and Doppler velocity time series from complex multi-line solar spectra by combining adaptive windowing, asymmetric Voigt options, and split-core handling, outperforming standard fast estimators on synthetic benchmarks.
Argus enables backdoor detection in decentralized ML by collaborative neighbor-based validation of triggers, backed by convergence theory and reducing attack success by up to 90% on tested datasets.
A reusable framework generates verification instances with provably known robustness labels, revealing numeric tolerance issues and bugs in five verifiers while introducing difficulty profiles to diagnose failure modes.
QLL is a novel logic for neuro-symbolic learning that uses ML-native operations (sum, log-sum-exp) on logits to embed constraints, satisfying most linear logic properties and showing stronger correlation between empirical robustness and formal verification than prior approaches.
Gradient matching empirically recovers implicit regularization effects such as l2 penalties from early stopping and dropout in neural networks.
A framework quantifies DNN complexity via tensor operations, links 40 years of breakthroughs to complexity increases, and releases a dataset of 3000+ unexplored high-complexity architectures.
Introduces Calibrated Size Ratio (CSR) and confidence-weighted metrics to better detect overconfidence risk and calibration issues beyond the limitations of ECE.
BRIDGE creates the first formal heterogeneous multi-dataset benchmark for IoT botnet detection with LODO evaluation, and TCH-Net achieves mean LODO F1 of 0.5577 while reaching F1 0.8296 on standard tests, outperforming twelve baselines.
SecurePix uses FeFET multidomain polarization states for in-pixel symmetric-key encryption, dropping ResNet-18 accuracy to 9.58% on MNIST and 6.98% on CIFAR-10 while supporting key-based decryption via lookup table.
S2-WEF detects dynamic free-riders in federated learning by simulating attack WEF patterns from prior global models, combining them with mutual deviation scores, and using two-dimensional clustering without proxy data or pre-training.
A multi-mode quantum annealing approach enables VAEs with Boltzmann priors, showing faster training and better generation than Gaussian-prior VAEs on MNIST, Fashion-MNIST, and CelebA plus improved out-of-distribution detection.
Shape- and peak-sensitive goodness functions for Forward-Forward deliver up to 72pp gains over sum-of-squares, reaching 98.2% on MNIST and 89% on Fashion-MNIST.
Langevin sampling on the modern Hopfield energy produces training-free stochastic attention that transitions from exact retrieval to generation as temperature rises, with an entropy inflection condition marking the shift.
A programmable superconducting LIF neuron with intrinsic static memory and dual-timescale plasticity achieves 45 GHz operation and femtojoule energy per spike.
Harder classification tasks produce neural representations whose accuracy collapses under binarization and shuffling while easier tasks remain robust, defining task complexity via the performance gap between full-precision and perturbed networks.
Sparse autoencoders resolve superposition in image-based neuron representations, recovering geometric fidelity and enabling scRNA-seq adaptation plus GW-map alignment to reconstruct pathology pathways without spatial transcriptomics.
Hessian eigenvector displacement and inverse participation ratio metrics show SGD stabilizing leading curvature directions while Adam causes more reorganization and parameter localization in MLP training.
Introduces Hell-Char, PaLit-Char and Med-Char datasets plus a similarity-weighted supervised contrastive loss and lacuna augmentation to learn diachronic embeddings for ancient Greek letterforms.
Local 2- and 3-cycles enhance RNN computational capacity for Boolean functions, predicted by structural statistics, while adding interneurons boosts large networks.
citing papers explorer
-
STRABLE: Benchmarking Tabular Machine Learning with Strings
A new corpus of 108 mixed string-numeric tables shows that advanced tabular learners with basic string embeddings perform well on most real-world data, while large LLM encoders help on free-text heavy tables.
-
Self-Organized Conformal Prediction: Reducing Regional Coverage Gaps with Unsupervised Group Discovery
SOCP uses self-organizing maps for unsupervised group discovery to enable local calibration in conformal prediction, reducing regional coverage gaps on benchmarks with small set-size increases while preserving validity guarantees.
-
Dolph2Vec: Self-Supervised Representations of Dolphin Vocalizations
Dolph2Vec is the first species-specific self-supervised model for dolphin vocalizations, trained on longitudinal recordings from five dolphins, that outperforms general baselines on signature whistle classification and detection while producing embeddings aligned with known whistle categories.
-
Inverse Critical Experiment Design via Gradient Optimization and a Multigroup Attention-Based Neural Network Architecture
A U-Net surrogate with multigroup attention pooling is trained on OpenMC sensitivity data and combined with gradient optimization to generate grid-based critical experiment geometries that achieve c_k values up to 0.97757 for HALEU fuel validation.
-
DELOS: Detecting Shallow Transits in Kepler Photometry Using a Contrastive-Learning Framework
DELOS applies contrastive learning to phase-folded light curves to detect shallow intermediate-to-long period transits, reporting 15.5% and 11.25% gains in combined precision-recall over BLS and TLS in low-SNR tests plus 3-80x speedups.
-
Adaptive multi-line fitting for stable line-core intensity and Doppler velocity
LineFit delivers more stable line-core intensity and Doppler velocity time series from complex multi-line solar spectra by combining adaptive windowing, asymmetric Voigt options, and split-core handling, outperforming standard fast estimators on synthetic benchmarks.
-
Your Neighbors Know: Leveraging Local Neighborhoods for Backdoor Detection in Decentralized Learning
Argus enables backdoor detection in decentralized ML by collaborative neighbor-based validation of triggers, backed by convergence theory and reducing attack success by up to 90% on tested datasets.
-
Stress-Testing Neural Network Verifiers with Provably Robust Instances
A reusable framework generates verification instances with provably known robustness labels, revealing numeric tolerance issues and bugs in five verifiers while introducing difficulty profiles to diagnose failure modes.
-
Quantitative Linear Logic for Neuro-Symbolic Learning and Verification
QLL is a novel logic for neuro-symbolic learning that uses ML-native operations (sum, log-sum-exp) on logits to embed constraints, satisfying most linear logic properties and showing stronger correlation between empirical robustness and formal verification than prior approaches.
-
Estimating Implicit Regularization in Deep Learning
Gradient matching empirically recovers implicit regularization effects such as l2 penalties from early stopping and dropout in neural networks.
-
On the Architectural Complexity of Neural Networks
A framework quantifies DNN complexity via tensor operations, links 40 years of breakthroughs to complexity increases, and releases a dataset of 3000+ unexplored high-complexity architectures.
-
Beyond ECE: Calibrated Size Ratio, Risk Assessment, and Confidence-Weighted Metrics
Introduces Calibrated Size Ratio (CSR) and confidence-weighted metrics to better detect overconfidence risk and calibration issues beyond the limitations of ECE.
-
BRIDGE and TCH-Net: Heterogeneous Benchmark and Multi-Branch Baseline for Cross-Domain IoT Botnet Detection
BRIDGE creates the first formal heterogeneous multi-dataset benchmark for IoT botnet detection with LODO evaluation, and TCH-Net achieves mean LODO F1 of 0.5577 while reaching F1 0.8296 on standard tests, outperforming twelve baselines.
-
Lightweight True In-Pixel Encryption with FeFET Enabled Pixel Design for Secure Imaging
SecurePix uses FeFET multidomain polarization states for in-pixel symmetric-key encryption, dropping ResNet-18 accuracy to 9.58% on MNIST and 6.98% on CIFAR-10 while supporting key-based decryption via lookup table.
-
Dynamic Free-Rider Detection in Federated Learning via Simulated Attack Patterns
S2-WEF detects dynamic free-riders in federated learning by simulating attack WEF patterns from prior global models, combining them with mutual deviation scores, and using two-dimensional clustering without proxy data or pre-training.
-
Multi-Mode Quantum Annealing for Generative Representation Learning with Boltzmann Priors
A multi-mode quantum annealing approach enables VAEs with Boltzmann priors, showing faster training and better generation than Gaussian-prior VAEs on MNIST, Fashion-MNIST, and CelebA plus improved out-of-distribution detection.
-
Selectivity and Shape in the Design of Forward-Forward Goodness Functions
Shape- and peak-sensitive goodness functions for Forward-Forward deliver up to 72pp gains over sum-of-squares, reaching 98.2% on MNIST and 89% on Fashion-MNIST.
-
Stochastic Attention via Langevin Dynamics on the Modern Hopfield Energy
Langevin sampling on the modern Hopfield energy produces training-free stochastic attention that transitions from exact retrieval to generation as temperature rises, with an entropy inflection condition marking the shift.
-
Programmable superconducting neuron with intrinsic in-memory computation and dual-timescale plasticity for ultra-efficient neuromorphic computing
A programmable superconducting LIF neuron with intrinsic static memory and dual-timescale plasticity achieves 45 GHz operation and femtojoule energy per spike.
-
Task complexity shapes internal representations and robustness in neural networks
Harder classification tasks produce neural representations whose accuracy collapses under binarization and shuffling while easier tasks remain robust, defining task complexity via the performance gap between full-precision and perturbed networks.
-
Resolving superposition in AI for interpretability and cross-modal alignment in patient-neuronal images
Sparse autoencoders resolve superposition in image-based neuron representations, recovering geometric fidelity and enabling scRNA-seq adaptation plus GW-map alignment to reconstruct pathology pathways without spatial transcriptomics.
-
Characterizing Optimizer-Dependent Training Dynamics Through Hessian Eigenvector Displacement and Localization
Hessian eigenvector displacement and inverse participation ratio metrics show SGD stabilizing leading curvature directions while Adam causes more reorganization and parameter localization in MLP training.
-
Learning Diachronic Representations of Ancient Greek Letterforms
Introduces Hell-Char, PaLit-Char and Med-Char datasets plus a similarity-weighted supervised contrastive loss and lacuna augmentation to learn diachronic embeddings for ancient Greek letterforms.
-
Identifying structural design principles shaping the computational abilities of recurrent neural networks
Local 2- and 3-cycles enhance RNN computational capacity for Boolean functions, predicted by structural statistics, while adding interneurons boosts large networks.
-
NASDAQ: Normalized Observation Space Dynamics-Augmented Q-Learning
NASDAQ normalizes observations in an online RL setting so that dynamics prediction losses are balanced across dimensions, yielding competitive performance with lower wall-time than prior model-based and self-predictive methods.
-
Another Look at Log-PCA for Probability Measures: A Dynamical Formulation and Statistical Convergence
Introduces Wasserstein Tangential PCA (WT-PCA) as a variational dynamical approach to log-PCA on the Wasserstein space and derives its empirical statistical convergence rate in 2-Wasserstein distance.
-
A Spectral Phase Diagram for Binary Few-Shot Classification: Intrinsic Dimensionality, Geometric Saturation, and Representational Diagnosis
Defines saturation index S(K) = erank(Σ̂_W^(K))/K that identifies when linear discriminant stabilizes in binary few-shot classification, with empirical phase diagram and stopping-rule AUC of 0.752 on 17 tasks.
-
Non-linear mechanical field reconstruction coupling recurrent neural networks with physics-informed graph neural networks
A coupled LSTM-GNN model reconstructs local elasto-plastic stress fields from macroscopic loading paths on a plate-with-hole microstructure, achieving 1000x speedup and mesh transferability with 1.9% error.
-
Direct High-Magnetic-Field Coupling to Stripe Order in a Cuprate Superconductor
High magnetic fields directly enhance the amplitude and correlation length of stripe order in a cuprate superconductor far above the vortex melting transition, indicating a coupling mechanism independent of superconductivity suppression.
-
What changes after deployment? A survey on On-device Learning in TinyML
A survey of on-device learning in TinyML organized by distribution change regimes, highlighting influences on applications, hardware, and solutions plus a gap between benchmarks and deployments.
-
Worker Disagreement Reveals Sharp Directions in Local SGD
Worker-average gaps in Local SGD serve as a Hessian-free estimator of the dominant sharp subspace by capturing gradient alignment with high-curvature directions.
-
On the Equivariant Learning of the $Q$-tensor Order Parameter
Equivariant neural networks for 2D Q-tensor prediction in nematic liquid crystals achieve lower errors and better generalization than non-equivariant models while satisfying symmetry constraints.
-
MEDAL: Manifold Embedding Distillation via Autoencoder Learning
MEDAL distills manifold embeddings into autoencoders to enable out-of-sample extension and held-out validation of dimension reduction methods.
-
Encrypted Neural Networks without Overflows
Introduces formal verification to compute certified neuron range bounds for CKKS-encrypted neural networks, eliminating overflow failures that previously reached 47%.
-
Expectation Consistency Loss: Rethink Confidence Calibration under Covariate Shift
Derives expectation consistency condition as necessary and sufficient for calibration under covariate shift and proposes ECL loss with matching sample complexity to ECE.
-
Generative Recursive Reasoning
GRAM is a latent-variable generative model that performs recursive reasoning via stochastic trajectories, trained with amortized variational inference to support multi-hypothesis reasoning and unconditional generation.
-
Aligned Training: A Parameter-Free Method to Improve Feature Quality and Stability of Sparse Autoencoders (SAE)
Aligned training reparameterizes SAEs to enforce unit alignment between encoder and decoder directions, yielding Pareto gains on SAEBench while removing dead features and improving stability.
-
The Diffusion Encoder
A diffusion model serves as the encoder in an autoencoder when trained alternately with the decoder to resolve opposing update directions while retaining the standard diffusion training objective.
-
From Clever Hans to Scientific Discovery: Interpreting EEG Foundational Transformers with LRP
LRP on EEG transformers reveals Clever Hans artifacts in motor imagery tasks and a recurring central electrode cluster as a candidate sensorimotor signature of arousal.
-
Instructions Shape Production of Language, not Processing
Instructions trigger a production-centered mechanism in language models, with task-specific information stable in input tokens but varying strongly in output tokens and correlating with behavior.
-
Inducing Spatial Locality in Vision Transformers through the Training Protocol
CutMix augmentation during training induces spatial locality in early layers of Vision Transformers trained from scratch, as measured by reduced Mean Attention Distance.
-
What If We Let Forecasting Forget? A Sparse Bottleneck for Cross-Variable Dependencies
MS-FLOW uses a capacity-limited sparse routing mechanism to model only critical inter-variable dependencies in time series data, achieving state-of-the-art accuracy on 12 benchmarks with fewer but more reliable connections.
-
Flow Matching with Arbitrary Auxiliary Paths
AuxPath-FM extends flow matching to arbitrary auxiliary distributions while preserving the continuity equation and marginal training objective.
-
P-Guide: Parameter-Efficient Prior Steering for Single-Pass CFG Inference
P-Guide achieves single-pass classifier-free guidance in flow matching by modulating the initial latent state and is equivalent to standard CFG under a first-order approximation while cutting latency by half.
-
When AI Meets Science: Research Diversity, Interdisciplinarity, Visibility, and Retractions across Disciplines in a Global Surge
AI use in science has grown exponentially since 2015 but stays confined to computer science and statistics topics, shows higher retraction rates and citations, and follows distinct global adoption patterns.
-
Calculating Domain of Attraction Boundary of Power Systems Based on the Gentlest Ascent Dynamics
Applies gentlest ascent dynamics and stable manifold methods to compute domain of attraction boundaries for stable equilibria in synchronous-generator power system models.
-
Class Angular Distortion Index for Dimensionality Reduction
CADI quantifies the preservation of relative cluster angles in low-dimensional projections using internal angles from point triples.
-
Empirical Insights of Test Selection Metrics under Multiple Testing Objectives and Distribution Shifts
A broad empirical benchmark shows how 15 existing test selection metrics perform for fault detection, performance estimation, and retraining under corrupted, adversarial, temporal, natural, and label shifts across image, text, and Android data.
-
Modulation Feature Enhancement with a Multi-Stage Attention Network for Underwater Acoustic Target Recognition
A 1-D CNN with novel multi-stage spectral attention mechanisms and adjustable class-balanced focal loss improves recognition accuracy on real ship-radiated noise datasets.
-
LTBs-KAN: Linear-Time B-splines Kolmogorov-Arnold Networks
LTBs-KAN delivers linear-time B-spline evaluation in KANs plus parameter reduction via product-of-sums factorization, with competitive results on MNIST, Fashion-MNIST, and CIFAR-10.