Skip to content
View addo561's full-sized avatar
  • 12:09 (UTC -12:00)

Block or report addo561

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
addo561/README.md

Korli Larry Addo

Generative AI · Diffusion Systems · Building toward Large World Models

profile views


What I'm building

I'm working toward a single long-term research goal: generative models that construct and reason about persistent, interactive 3D worlds. I'm building the foundation systematically — one layer at a time, from diffusion internals to spatial intelligence.

The arc:

  • 🔬  Now — Diffusion Architecture: Replacing U-Nets with scratch-built Diffusion Transformers (DiT). Mapping GFLOPs scaling behavior across resolutions. Understanding the math before touching the API.
  • 🎬  Next — Video Generation: Cross-frame attention for temporal consistency. Fighting flickering directly. Evaluating with FVD, not vibes.
  • 🧊  Then — 3D Spatial Generation: Score Distillation Sampling over NeRF / 3D Gaussian Splatting. Documenting gradient flow strictly.
  • 🌍  Goal — Large World Models: Action-conditioned generative systems that build coherent, navigable 3D environments.

Every project ships with a technical report. Results are reported honestly, including failures.


Stack

Core ML

Python PyTorch NumPy OpenCV

Tooling

FastAPI Git


GitHub Stats

GitHub trophies

Top languages


Connect

X / Twitter   LinkedIn   Email

Pinned Loading

  1. Conditional-Variational-AutoEncoder-cvae- Conditional-Variational-AutoEncoder-cvae- Public

    A PyTorch implementation of a Conditional Variational Autoencoder (CVAE) designed for controlled data generation.

    Jupyter Notebook

  2. DDIM-With-CFG DDIM-With-CFG Public

    DDIM with classifier free guidance

    Jupyter Notebook

  3. engine-2-transformer engine-2-transformer Public

    Transformer decoder built from scratch following Sebastien Raschka's approach — tokenization through to autoregressive text generation, no training frameworks.

    Jupyter Notebook

  4. mml-book.github.io mml-book.github.io Public

    Forked from mml-book/mml-book.github.io

    Companion webpage to the book "Mathematics For Machine Learning"

    Jupyter Notebook