hypogenic - Generate and test hypotheses from datasets
Automates hypothesis generation and testing from empirical datasets, with optional integration of research literature.
Tags
Updated: 2026-10-16Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Generate data-driven hypotheses
- Test competing hypotheses
- Integrate literature insights
- Refine hypotheses iteratively
- Process research PDFs
- Cache repeated LLM responses
Inputs
- Dataset files
- Task configuration YAML
- Prompt templates
- Research paper PDFs
- LLM provider or local model
- Label extraction function
Outputs
- Generated hypothesis bank files
- Hypothesis inference result files
- Processed literature files
Requirements
- Python environment with hypogenic installed
- HuggingFace-format dataset files
- API-based or local LLM support
- GROBID for PDF preprocessing
- s2orc-doc2json for literature processing