Skip to content

Competition Solution · Python

simon-arc-lab

My solution for ARC-Prize 2024

What it is

This repository is a solution for the ARC Prize 2024 competition, implementing multiple approaches (LLM-based and decision tree-based) to solve abstract reasoning tasks using RLE-compressed puzzle representations. The committed state includes model code, dataset generation tools, and training notebooks, though the notebooks have not been executed to demonstrate training or evaluation results.

  • arc-agi
  • arc-prize-2024
Scored atd8a262114 Nov 2026Changes since

Merit describes the code at this commit. Later pushes count once the owner rescans. How merit is scored

Distribution based on AI work

1 / 2
Machine Learning: 27.3%Computer Vision: 24.2%NLP & LLMs: 21.2%Data Science: 18.2%MLOps: 6.1%AI Agents: 3.0%
  • Machine Learning27%
  • Computer Vision24%
  • NLP & LLMs21%
  • Data Science18%
  • MLOps6%
  • AI Agents3%

Each field's share of the project's AI work, as scoring counts it. It ranks in the field it is most about.

What the judge found

Read from the project's code at its scored commit.

Strengths3

  • Original RLE compression approach for representing ARC puzzles in LLMs
  • Comprehensive dataset generation system with multiple transformation techniques
  • Exploration of multiple modeling approaches (LLM and decision tree)

Limitations

The owner keeps this project's limitations private.

Around it on the Machine Learning board

Whole board
  1. 99Linear-Regression-ScratchEducational Implementation · Jupyter Notebook50.0/100Implementationkamran akmal
  2. 100, shared rankImage-Classification-CIFAR100ML Application · Python49.7/100Implementationkamran akmal
  3. 100, shared rankYOLOvRestNetvMobileNetTraining Experiment · Python49.7/100ImplementationDaffa
  4. 100, shared ranksimon-arc-labCompetition Solution · Python49.7/100ImplementationSimon Strandgaard
  5. 103rife-tradeML Application · TypeScript49.5/100ImplementationPranav

More from Simon Strandgaard