Contrastive learning operates on a simple yet effective principle: Embeddings of positive pairs are pulled together, while those of negative pairs are pushed apart. In this paper, we propose a unified framework for understanding contrastive learning...
A framework bringing additional nodes & filters to Stencil dedicated to Swift code generation (⭐ 302)
3 days ago · Ellen DeGeneres enjoyed decades of success until a viral tweet calling her one of the "meanest people alive" sparked allegations and her fall from frame.
Jersey number recognition is an important task in sports video analysis, partly due to its importance for long-term player tracking. It can be viewed as a variant of scene text recognition. However, there is a lack of published attempts to apply scen...
Corpus distillation for biomedical large language models (LLMs) seeks to address the pressing challenge of insufficient quantity and quality in open-source annotated scientific corpora, which remains a bottleneck for effective LLM training in biomedi...
This study introduces a hypothesis-testing framework to assess whether large language models (LLMs) possess genuine reasoning abilities or primarily depend on token bias. We go beyond evaluating LLMs on accuracy; rather, we aim to investigate their t...
This paper presents a new framework for jointly enhancing the resolution and the dynamic range of an image, i.e., simultaneous super-resolution (SR) and high dynamic range imaging (HDRI), based on a convolutional neural network (CNN). From the common...
Documenting the construction of an NMT (Neural Machine Translation) system for En/Ja based on the Transformer architecture leveraging the OpenNMT framework. A systematic exploration of corpora pre-processing, hyperparameter tuning and model architect...
Lattice distortions in which the axes are locally rotated provide an intrinsic source of frustration in anisotropic superconductors. A general framework to study this effect is presented. The influence of lattice defects and phonons in $d$ and $s+d...
Collision-tolerant trajectory planning is the consideration that collisions, if they are planned appropriately, enable more effective path planning for robots capable of handling them. A mixed integer programming (MIP) optimization formulation demons...
Dark Matter searches in collider and non-collider experiments requires systematic and consistent approach. We suggest and perform classification of Minimal Consistent Dark Matter models which are aimed to create a solid framework for Dark Matter expl...
Studies Beyond the Standard Model (BSM) will become more and more important in the near future with a rapidly increasing amount of data from different experiments around the world. The full study of BSM models is in general an extremely time-consumin...
Theoretical calculations Beyond the Standard Model (BSM) constitute a challenge for high energy physicists, but are necessary when searching for New Physics. The predictions of a BSM scenario need to be compared with experimental data and the Standar...
We propose NH-TTC, a general method for fast, anticipatory collision avoidance for autonomous robots having arbitrary equations of motions. Our proposed approach exploits implicit differentiation and subgradient descent to locally optimize the non-co...
KL's Portfolio v2. Made with Love using NextJS, ChakraUI and Framer motion. (⭐ 94)
We review in a tutorial fashion some of the causes of impurity density variations along field lines and radial impurity transport in the moment approach framework. An explicit and compact form of the parallel inertia force valid for arbitrary toroida...
We present a suite of high-resolution numerical simulations to study the evolution and survival of dust in hot galactic winds. We implement a novel dust framework in the Cholla hydrodynamics code and use wind tunnel simulations of cool, dusty clouds...
X-ray transmission windows for the low energy range, especially between 0.1 keV and 1 keV have been designed and fabricated based on graphenic carbon (GC) with an integrated silicon frame. A hexagonal and a bar grid support structure design have been...
Sep 2, 2025 · Throughput is a measure of how much data or work can be processed within a given time frame, making it a critical performance metric in networking and IT. Optimizing throughput improves …
A full stack 'Hacker News' style demo web application built with React + Redux (Front) and django REST Framework (Server). (⭐ 135)