Skip to content

knowledge-base-rag-preparation.md

Latest commit

 

History

History
14 lines (9 loc) · 932 Bytes

knowledge-base-rag-preparation.md

File metadata and controls

14 lines (9 loc) · 932 Bytes
Raw

Knowledge Base / RAG Preparation

Use this category for examples where autoresearch collects, structures, cleans, or refreshes information for retrieval, memory, or knowledge system inputs.

Submission format

- [Name](URL) - Industry: one-sentence description of the autoresearch use case.

Entries

  • AutoRAGsearch - RAG retrieval optimization: applies an autoresearch-style loop to a fixed QA benchmark by editing only rag_pipeline.py, running local retrieval experiments, and improving retrieval_score from 0.9472 to 0.9867 over 20 autonomous experiments.
  • AutoRAG Research Lab - RAG pipeline optimization: runs autonomous agents that plan, implement, benchmark, and keep/discard retrieval pipeline changes using Karpathy-mode over EnterpriseRAG-Bench with Docker-sandboxed evaluation and semantic no-op detection.