11971 papers
This paper presents an automated, multi-stage pipeline that identifies cervical spine fractures by fusing orthogonal 2D segmentations to estimate 3D volumes of interest, which are then analyzed using a 2.5D CNN-Transformer ensemble to achieve diagnostic performance comparable to expert radiologists while reducing computational dimensionality.
This paper establishes a unified statistical framework demonstrating that synthetic augmentation in imbalanced learning is not universally beneficial, revealing that its efficacy and optimal quantity depend on local data symmetry and generator alignment, and proposing a Validation-Tuned Synthetic Size (VTSS) strategy to empirically determine the best augmentation level.
SpecBridge is a novel framework that improves small-molecule identification from mass spectrometry by fine-tuning a spectral encoder to align with a frozen molecular foundation model's latent space, achieving significant accuracy gains over existing baselines while maintaining parameter efficiency.
This paper presents LeanTutor, a proof-of-concept AI system that combines Large Language Models with the Lean theorem prover to provide verified mathematical proof tutoring, evaluated using the newly introduced PeanoBench dataset.
This paper introduces NRR-Phi, a formal framework that maps ambiguous text to a non-collapsing state space using a hybrid extraction pipeline, thereby preserving multiple interpretations and preventing premature semantic commitment in large language model inference.
This paper proposes a new deployment paradigm for language models called "Least-Privilege Language Models," which introduces a mechanism to dynamically restrict a model's internal computational capabilities during inference—rather than just filtering outputs—thereby enabling fine-grained, policy-driven control over specific functionalities without the need for retraining.
NeuroPareto is a novel framework that integrates rank-centric filtering, uncertainty disentanglement via Deep Gaussian Processes, and a history-conditioned acquisition network to efficiently navigate costly many-objective optimization problems in vast parameter spaces while outperforming existing baselines in convergence and diversity.
This paper proposes a consensus-based Bayesian framework that models enterprise directory access as a multi-level interaction graph to detect malicious activity by identifying logical perturbations in strongly connected components through opinion dynamics and time-evolving anomaly scoring.
The paper proposes HealthMamba, an uncertainty-aware spatiotemporal graph state space model that integrates a unified context encoder, a novel GraphMamba architecture, and comprehensive uncertainty quantification to significantly improve the accuracy and reliability of healthcare facility visit predictions across four large-scale real-world datasets.
This paper proposes "Rewards as Labels" (REAL), a novel framework that reformulates Reinforcement Learning with Verifiable Rewards as a classification problem to address gradient misassignment and domination issues in methods like GRPO, thereby achieving superior training stability and performance on mathematical reasoning benchmarks compared to state-of-the-art baselines.
The paper introduces SureLock, a method for Masked Diffusion-LMs that dynamically locks converged token positions to skip redundant computations while caching their keys and values, thereby reducing per-iteration complexity from to $O(MNd)$ and achieving 30–50% FLOP savings without compromising generation quality.
This paper extends the CNF framework to solve coupled and non-linear Partial Differential Equations for forward, inverse, and discovery tasks, while providing a comprehensive survey, implementation, and evaluation of various neural PDE solvers to address challenges like the curse of dimensionality and high computational costs.
This paper establishes a homology theory for ample groupoids using compactly supported Moore complexes, proving functoriality and Kakutani invariance, deriving Mayer-Vietoris sequences, and demonstrating that a universal coefficient theorem holds for discrete coefficients while identifying specific obstructions for non-discrete ones.
FastLSQ is a high-performance framework that solves linear and nonlinear partial differential equations (PDEs) and inverse problems with exceptional speed and accuracy by leveraging sinusoidal random Fourier features to enable exact, closed-form analytical derivatives, thereby eliminating the need for automatic differentiation and iterative training.
The First International StepUP Competition leveraged the newly released UNB StepUP-P150 dataset to advance biometric footstep recognition, culminating in a global contest where the top team achieved a 10.77% equal error rate while highlighting persistent challenges in generalizing to unfamiliar footwear.
This paper introduces TIME, a next-generation, task-centric benchmark comprising 50 fresh datasets and 98 forecasting tasks that addresses critical limitations in existing evaluations by ensuring data integrity, aligning with real-world requirements, and proposing a novel pattern-level perspective to rigorously assess the zero-shot generalization capabilities of time series foundation models.
Chimera is a neuro-symbolic framework that enables trustworthy, line-rate traffic analysis on programmable dataplanes by mapping attention-based neural computations and symbolic constraints onto hardware primitives to achieve high-fidelity inference within strict resource limits.
This paper proposes JPmHC, a novel framework that stabilizes and scales Hyper-Connections by replacing identity skips with trainable linear mixers constrained to operator-norm-bounded manifolds, thereby preserving Jacobian spectra for improved training stability, memory efficiency, and performance on tasks like ARC-AGI.
The paper introduces DRESS, a deterministic and parameter-free framework that generates isomorphism-invariant graph fingerprints via a convergent non-linear dynamical system, offering expressiveness comparable to the 2-WL test at significantly lower computational cost while generalizing to enhanced variants like -DRESS that surpass higher-order Weisfeiler-Leman tests.
FlowCorrect is a modular, interactive imitation learning framework that enables real-time, sample-efficient adaptation of generative flow-matching robotic policies through sparse human pose corrections, significantly improving success rates on previously failed tasks without retraining the underlying model.