Kwanwoo (Eric) Kim

Every note, newest first.

• Eyes Wide Shut? Exploring the Visual Shortcomings of Multimodal LLMs • The study investigates the visual limitations of multimodal large language models (MLLMs), specifically focusing on visual encoders like CLIP.Jan 22 25
• Prometheus: Inducing Fine-grained Evaluation Capability in Language Models • Originates from the growing reliance on LLMs for evaluating human-level text generation.Jan 21 25
• LORA: LOW-RANK ADAPTATION OF LARGE LANGUAGE MODELS • Many applications in natural language processing rely on adapting one large-scale, pre-trained language model to multiple downstream applications.Jan 14 25
• Assessing Multimodal LLM-as-a-Judge with Vision-Language Benchmark • Can MLLMs effectively serve as judges in the multimodal domain, and how closely do their evaluations align with human preferences?Jan 11 25
• Latent Dirichlet Allocation - Part.1 • LDA's main function is not dimension reduction but Topic Modeling, and the definition of Topic Modeling, borrowing from the expression in the paper, is as followsDec 07 23
• From Seq2Seq to Transformer - Part.2 • The transformer model represents another significant leap in the NLP field after the Seq2Seq model and the attention mechanism.Oct 07 23
• From Seq2Seq to Transformer - Part.1 • The attention mechanism was proposed to improve the Seq2Seq model. To understand the attention mechanism well, you must understand the Seq2Seq model.Oct 04 23
• Generative AI - Part.2 • The learning and generation principles of Diffusion Models, which model multi-layer transformations and inverse transformations between data distribution and noise distribution, share many similarities with Flow Models.Jul 13 23
• Generative AI - Part.1 • So, on what principle do generative models, the core of generative AI, learn and operate? What does 'generation' mean? In the fields of statistics or machine learning, generation is defined as sampling from a probability distribution in data space, which was not observed or used during the learning process.Jul 12 23
• How Ray Tracing works Part.1 • Today we'll look at <Ray tracing> the dominant technique used in the graphics industry.Oct 02 22
• Image Style Transfer (CNN) • Paper Review of Image Style Transfer Using Convolutional Neural Networks.Jan 29 22
• 02. MLOps: Model Monitoring - Weights and Bias • Basics of MLOps using basic vision project.Jan 23 22
• Physics-based Human Motion Estimation and Synthesis from Videos • Paper review of Physics-based Human Motion Estimation and Synthesis from VideosJan 19 22
• 01. MLOps: Project Settings • Basics of MLOps using basic vision project.Jan 16 22
• Gradient Descent • Gradient descent method is also called as steepest descent method.Oct 16 21
• Linear Regression • Linear regression is a linear approach for modeling the relationship between a scalar response and one or more explanatory variables.Oct 09 21
• Adaptive Moment Estimation (ADAM) • ADAM is the use of Gradient Descent with Momentum and RMSProp at the same time.Oct 09 21
• KL Divergence • KL Divergence (Kullback-Leibler) also known as Relative Entropy is all about the difference between two distributions.Sep 27 21
• Bayesian Probability • Bayesian probability is an interpretation of the concept of probability.Sep 26 21
• Vanilla Generative Adversarial Networks (GAN) Review • Framework for estimating generative models.Sep 24 21