Reranking Lab

Retrieve broadly with Hybrid RAG, then refine candidate ordering with a cross-encoder before selecting the final context for generation.

GuidedNo API keys in Lab

Guided execution

This Lab replays a recorded Cookbook execution in your browser — no API keys required here. Running the Cookbook locally uses your own provider credentials.

Example question

Pick one question, then run the recorded pipeline.

Is calendar certificate rotation the fix for E_CONN_42?Intermittent TLS handshake failures without a stable error code — checklist?Should I keep using getUserProfile or switch for concurrent UIs?

Execution

Choose a question and run the recorded pipeline to retrieve broadly, rerank with a cross-encoder, then select the final context.

You will see: Chunking → Embeddings → BM25 → Dense → Lexical → RRF → Cross-encoder → Prompt → LLM → Answer

Previous Lab

Previous: Hybrid RAG — dense + lexical retrieval with RRF.

Continue Learning

Deeper explanations in the DataAIHub knowledge graph.

Implementation

Run the same pipeline locally from the public Cookbook.

Related Tools

Canonical tools often used with retrieval and reranking stacks.

Lab ID reranking · guided / precomputed