arXiv subjects

Mingi Kang

Publications and source records attributed to Mingi Kang.

5 recordsLinked to original sources

arXiv2026

Interpolation between Convolution and Attention via K-Nearest Neighbors

The shift from Convolutional Neural Networks to Transformers has reshaped computer vision, yet these two architectural families are typically viewed as fundamentally distinct. Convolutional Neural Networks are defined by spatially local convolution operations, while Transformers rely on global self-attention. We argue that convolution and self-attention, despite their apparent differences, can be unified within a single k-nearest neighbor aggregation framework. The critical insight is that both operations are special cases of neighbor selection and weighted aggregation. Convolution selects neighbors by spatial proximity while self-attention selects by feature similarity, revealing that they lie on a continuous spectrum rather than representing categorically different computations. We introduce Convolutional Nearest Neighbors (ConvNN), a unified framework that formalizes this connection. ConvNN exactly recovers standard and depthwise convolution by restricting neighbor selection to normalized spatial coordinates, and exactly recovers self-attention and its sparse variants, including KVT-attention, by replacing spatial proximity with scaled dot-product similarity. Beyond these special cases, ConvNN serves as a drop-in replacement for both convolution and attention layers, enabling systematic exploration of the intermediate spectrum between local and global aggregation through configurable similarity functions, neighbor selection strategies, positional encodings, and aggregation kernels.

cs.CV↗

arXiv2026

Insights into the Nature of Quantum Emitters in Electron-Irradiated hexagonal Boron Nitride

Quantum emitters in hexagonal boron nitride (hBN) have emerged as a promising solid-state platform for quantum technology applications. However, a persistent challenge in the field is the unclear origin of many observed emission lines, particularly in the visible range, which can be difficult to distinguish from signals arising from organic or process-induced contamination during sample preparations and handling. This ambiguity limits both the reproducibility of emitter generation and the reliable identification of truly intrinsic quantum defects. This work provides a step-by-step framework to assess whether quantum emitters in electron-irradiated hBN are associated with organic contaminants introduced during sample preparation. We employ hyperspectral imaging, thermal annealing, and oxygen plasma etching to investigate the origin of the green-yellow emitters in electron-irradiated hBN. The combined results not only rule out organic contamination as the source of emission but also provide insight into the spectral variability, thermal stability, and vertical localization of the emitters generated in electron-irradiated hBN that was created without any pre- or post-processing. In addition, our experiments demonstrate the feasibility of creating stable emitters in hBN with thicknesses below 10 nm. These findings provide practical guidance for the identification and controlled implementation of hBN-based single-photon emitters in quantum photonic devices.

physics.optics↗

arXiv2026

IGLU: The Integrated Gaussian Linear Unit Activation Function

Activation functions are fundamental to deep neural networks, governing gradient flow, optimization stability, and representational capacity. Within historic deep architectures, while ReLU has been the dominant choice for the activation function, modern transformer-based models increasingly are adopting smoother alternatives such as GELU and other self-gated alternatives. Despite their empirical success, the mathematical relationships among these functions and the principles underlying their effectiveness remains only partially understood. We introduce IGLU, a parametric activation function derived as a scale mixture of GELU gates under a half-normal mixing distribution. This derivation yields a closed-form expression whose gating component is exactly the Cauchy CDF, providing a principled one-parameter family that continuously interpolates between identity-like and ReLU-like behavior via a single sharpness parameter $\sigma$. Unlike GELU's Gaussian gate, IGLU's heavy-tailed Cauchy gate decays polynomially in the negative tail, guaranteeing non-zero gradients for all finite inputs and offering greater robustness to vanishing gradients. We further introduce IGLU-Approx, a computationally efficient rational approximation of IGLU expressed entirely in terms of ReLU operations that eliminates transcendental function evaluation. Through evaluations on CIFAR-10, CIFAR-100, and WikiText-103 across ResNet-20, ViT-Tiny, and GPT-2 Small, IGLU achieves competitive or superior performance on both vision and language datasets against ReLU and GELU baselines, with IGLU-Approx recovering this performance at substantially reduced computational cost. In particular, we show that employing a heavy-tailed gate leads to considerable performance gains in heavily imbalanced classification datasets.

cs.LG↗

arXiv2025

Parallel qMRI Reconstruction from 4x Accelerated Acquisitions

Magnetic Resonance Imaging (MRI) acquisitions require extensive scan times, limiting patient throughput and increasing susceptibility to motion artifacts. Accelerated parallel MRI techniques reduce acquisition time by undersampling k-space data, but require robust reconstruction methods to recover high-quality images. Traditional approaches like SENSE require both undersampled k-space data and pre-computed coil sensitivity maps. We propose an end-to-end deep learning framework that jointly estimates coil sensitivity maps and reconstructs images from only undersampled k-space measurements at 4x acceleration. Our two-module architecture consists of a Coil Sensitivity Map (CSM) estimation module and a U-Net-based MRI reconstruction module. We evaluate our method on multi-coil brain MRI data from 10 subjects with 8 echoes each, using 2x SENSE reconstructions as ground truth. Our approach produces visually smoother reconstructions compared to conventional SENSE output, achieving comparable visual quality despite lower PSNR/SSIM metrics. We identify key challenges including spatial misalignment between different acceleration factors and propose future directions for improved reconstruction quality.

cs.CV↗

arXiv2025

Unifying Convolution and Attention via Convolutional Nearest Neighbors

Convolutional Neural Networks and Vision Transformers are the two dominant architectural families in computer vision, defined by spatially local convolution and global self-attention respectively. Despite their apparent differences, we show that both operations are special cases of a single $k$-nearest neighbor aggregation framework: convolution selects neighbors by spatial proximity while attention selects by feature similarity, placing them at two ends of a shared operational spectrum. We introduce Convolutional Nearest Neighbors (ConvNN), a unified framework that exactly recovers standard and depthwise convolution, self-attention, and sparse attention variants including KVT-attention as special cases, and exposes the design space of neighbor-selection strategies between them through configurable similarity functions, positional encodings, and aggregation kernels. We validate ConvNN on ImageNet-1K classification across two complementary architectures: a hybrid branching layer in ResNet-50 that combines local and global feature learning, improving top-1 accuracy by 3.0% over the ResNet-50 baseline, and ConvNN-attention in ViT-Base that achieves 81.64% top-1 accuracy, surpassing standard multi-head self-attention by 0.7%. Together, these results demonstrate that ConvNN provides a principled foundation for designing operations that bridge convolutional and attention-based computation.

cs.CV↗