This project leverages a Large Language Model (LLM) to extract key settings and experiment results from an extensive volume of literature, enabling the design of experiments tailored to specific research targets
- Extract related sections from research papers with model1 (or keyword filter).
- Extract experiment setting and results from the selected section from model2.
- Create a dataset comprising experiment settings.
- Train/Fine-tune an inference model3 with the dataset, using it to design experiments
Llama index: https://github.com/jerryjliu/llama_index