4045 papers
The paper introduces HAPEns, a novel post-hoc ensembling method for tabular data that constructs diverse ensembles along the Pareto front of predictive performance and hardware efficiency, significantly outperforming existing baselines across 83 datasets by explicitly balancing accuracy with resource constraints.
This paper empirically demonstrates that contrary to the hypothesis that moral reasoning alignment requires diversity-seeking algorithms, standard reward-maximizing RLVR methods are equally or more effective because high-reward moral responses exhibit a concentrated distribution in semantic space similar to logical reasoning tasks.
This paper establishes a mathematical framework called Gradient Flow Drifting that proves the equivalence between the recently proposed Drifting Model and the Wasserstein gradient flow of the forward KL divergence under KDE approximation, while extending the approach to a mixed-divergence strategy on Riemannian manifolds to simultaneously mitigate mode collapse and blurring.
This technical note presents a JAX-compatible implementation of the Self-Scaled Broyden family of quasi-Newton methods, including BFGS, DFP, and Broyden variants with Zoom line search, built upon the Optimistix library to facilitate their adoption within the JAX community.
This paper introduces Geo-ATBench, a new benchmark and the Geo-AT task that leverage geospatial semantic context to resolve acoustic ambiguities in multi-label audio tagging, demonstrating through the GeoFusion-AT framework that incorporating location-based priors significantly improves recognition performance and aligns with human judgment.
This paper proposes Conditional Expectation Reward (CER), a novel reinforcement learning method that utilizes the large language model itself as an implicit verifier to provide soft, graded reward signals, thereby overcoming the limitations of rule-based verification and enabling effective reasoning training across both mathematical and general free-form answer domains.
This paper proposes a novel, explainable approach to detect and eliminate neural network backdoors by analyzing active paths within the model, demonstrating its effectiveness through experiments on intrusion detection systems.
The paper introduces FAME, a novel abductive explanation method for large neural networks that utilizes dedicated perturbation domains and LiRPA-based bounds to efficiently generate formal abstract minimal explanations, demonstrating superior performance in explanation size and runtime compared to VERIX+.
This paper proposes a Spatio-Temporal Attention Graph Neural Network (STA-GNN) that integrates conformal prediction to provide unsupervised, explainable, and drift-aware anomaly detection for Industrial Control Systems by modeling dynamic inter-dependencies across physical and network entities.
This paper demonstrates that a data-driven framework combining Singular Value Decomposition with Shallow Recurrent Decoder (SHRED) networks can accurately and efficiently reconstruct full spatio-temporal magnetohydrodynamic states from sparse temperature sensor measurements, offering a robust surrogate model for real-time monitoring and control in nuclear fusion applications.
This paper introduces Contract And Conquer (CAC), a black-box adversarial attack method that combines knowledge distillation on an expanding dataset with precise search space contraction to provably compute adversarial examples within a fixed number of iterations, outperforming existing state-of-the-art approaches on ImageNet.
This paper introduces EvoSchema, a comprehensive benchmark featuring a novel taxonomy of ten schema perturbation types to evaluate and enhance the robustness of text-to-SQL models against real-world database schema evolution, revealing that table-level changes significantly impact performance and demonstrating that training on diverse schema designs improves model resilience.
This paper introduces Riemannian MeanFlow (RMF), a novel framework that enables efficient one-step generative modeling on manifolds by defining an average-velocity field via parallel transport and utilizing a log-map tangent representation to avoid costly numerical integration while maintaining high sample quality.
This paper introduces "Sample-and-Search," a learning-augmented algorithm for high-dimensional -median clustering that utilizes a predictor to preprocess data, thereby significantly reducing both computational complexity and exponential dimensionality dependency while achieving lower clustering costs compared to state-of-the-art methods.
CacheSolidarity is a lightweight system that secures multi-tenant LLM serving against Automatic Prefix Caching side-channel attacks by selectively isolating suspicious cache reuse, thereby achieving significantly higher cache efficiency and lower latency compared to existing all-or-nothing isolation defenses.
This paper compares Monte Carlo Dropout and Conformal Prediction for uncertainty estimation in CNNs trained on Fashion-MNIST, revealing that while H-CNN VGG16 achieves higher accuracy, GoogLeNet offers better calibration and Conformal Prediction provides statistically guaranteed reliability for high-stakes applications.
This paper proposes a structural remedy for data leakage in machine learning by introducing a grammar of seven kernel primitives connected by a typed directed acyclic graph with four hard constraints, specifically a runtime-enforced terminal assess constraint that prevents selection and memorization leakage, a claim validated across 2,047 experimental instances and implemented in Python, R, and Julia.
CUPID is a novel, plug-in framework that enables joint estimation of aleatoric and epistemic uncertainty in pretrained deep learning models without requiring retraining or architectural modifications, thereby enhancing interpretability and trust in high-stakes AI applications.
This paper proposes a deep learning-based autoencoder architecture for the Randomized Distributed Function Computation (RDFC) framework that minimizes the total variation distance to an unknown target distribution using only data samples, demonstrating superior communication efficiency compared to traditional data compression methods, particularly under limited common randomness.
This paper proposes a method to protect intellectual property in neural network models by binding their weights to unique hardware properties using Physically Unclonable Functions (PUFs), thereby preventing accurate execution on cloned hardware.