Project Overview (Placeholder)
This page will introduce my work on a Retrieval-Augmented Generation (RAG) evaluation platform, including dataset construction, vector store settings, evaluation metrics, and LLM agent workflow.
🚧 Full content is coming soon…
Upcoming Sections (Placeholder)
Planned additions:
- System architecture diagram
- Semantic chunking vs. default chunking comparison
- Evaluation metrics (Correctness, Citation Precision, etc.)
- OpenSearch vs. Serverless Knowledge Base experiments
- Example prompts and outputs