survey

What problem does it solve?

This Skill automates the process of surveying the State of the Art (SOTA) literature for a given Artificial Intelligence/Machine Learning (AI/ML) topic, method, or architecture, providing actionable insights for implementation.

Core Features & Use Cases

• Literature Survey: Identifies and summarizes top research papers on a specified AI/ML topic.
• Comparative Analysis: Builds a comparison table of methods, including key ideas, benchmark results, compute requirements, and code availability.
• Implementation Recommendation: Recommends the best-suited method for the current codebase based on identified constraints and provides a concrete implementation plan.
• Use Case: When starting a new project involving "object detection for small objects," this Skill can quickly identify the most effective SOTA methods, compare their performance and resource needs, and suggest the best approach with a clear path to integration.

Quick Start

Use the survey skill to find the best methods for "efficient transformers" in our current PyTorch codebase.