This paper establishes a large deviation principle for the generalization error of interpolating classifiers in the overparametrized regime.

This paper investigates limitations of learning tanh neural networks under finite-precision computations and Lp accuracy guarantees.

This paper develops statistical learning theory for gradient boosting in Peaks-over-Threshold modeling using Generalized Pareto distributions, deriving error bounds and reducing gradient correlation.

The paper proposes WEINCE, a modified InfoNCE objective that uses extreme value theory corrections to improve contrastive learning by more accurately modeling the selection of hard negative examples.

The paper analyzes low-degree estimation thresholds for recovering hidden signals in planted hypergraphs and tensor PCA, establishing sharp phase transitions and providing polynomial-time recovery alg…

The paper introduces a robust, two-part framework (HyPE and HyPS) using hyperbolic geometry to efficiently detect and sanitize malicious prompts targeting Vision-Language Models (VLMs).

This paper introduces and analyzes a consistent estimator for the sub-Gaussian parameter ($\xi_*^2$), providing convergence rates and demonstrating its applicability in large-scale biological enrichme…

The paper argues that large activation spikes in LLMs are structural vector biases, and proposes a novel quantization framework (INSERTQUANT) to eliminate these spikes, enabling robust low-bit quantiz…

This paper proposes Tail-Aware Adaptive-k (TAA-k), a training-free framework for adaptive context selection in retrieval-augmented generation systems using Extreme Value Theory.

While restricting a model to the theoretical Markov boundary can significantly improve prediction, the practical process of discovering and using this boundary is often computationally infeasible and…

The paper proposes a Lagrangian sub-flow (LSF) framework and geometric diagnostic signals to improve out-of-distribution detection using Continuous Normalizing Flows, overcoming the likelihood paradox…

This paper investigates the application of Parameter-Efficient Fine-Tuning (PEFT) methods, specifically adapters and LoRA, to large pretrained models for instance segmentation, demonstrating that thes…

This paper develops a unified spectral analysis framework to explain how knowledge transfer (KT) works across different machine learning regimes, such as Knowledge Distillation and Weak-to-Strong gene…

This paper analyzes the multi-regime behavior of Scientific Machine Learning (SciML) models, finding that optimization effectiveness is regime-specific and that failure modes require a unified, regime…

The paper proposes sampling directly from approximations of an LLM posterior, conditioned on high-scoring regions, to generate more coherent and useful text compared to existing post-hoc hallucination…

This paper improves the theoretical bounds for estimating discrete probability distributions using the $\ell_\infty$ norm, resolving several open questions in the field.

This paper introduces BBOmix, an open-source benchmark for unsupervised representation learning on real-world biological data.

The paper formalizes the concept of calibration for probabilistic label ranking, demonstrating that popular models are often poorly calibrated and that calibration captures a meaningful quality dimens…

This paper introduces a mechanistic neuronal network model for multilayer learning, offering biological insights and an alternative to backpropagation.

The paper explains the 'table-chart gap' in scientific claim verification by showing that multimodal LLMs successfully encode information from charts but fail to route it to the final prediction layer…