Activity v1

Chunking and Reranking Laboratory

Compare boundary-aware chunks candidate breadth reranker strength context cost and retrieval quality under explicit assumptions.

completion
beginner
seed 11

Chunking and Reranking Laboratory: Baseline

Explore the core controls with stable inputs and visible assumptions.

Assumptions

The simulation is deterministic and intentionally simplifies provider and hardware behavior.

Failure injection

Stable baseline with no injected production fault.

Boundary coverage81%
Candidate recall84%
Final precision80%
Context budget2,520 tok
The pipeline has a workable candidate-to-context funnel; validate it on labeled queries before choosing these values.

Simplified deterministic model: values reveal trade-offs and are not production benchmarks.

0 saved attempts

Expected outcomes

  • Diagnose chunk-boundary and candidate-set failures
  • Balance recall precision and context cost

Connected concepts

Apply this lab in an architecture