Anam #06: "512-token chunks are too aggressive. A complex argument spanning 2000–4000 tokens gets cut. Evidence separated from conclusions."
Proposed: Large chunks (2,048 tokens) for Stage 2 extraction. Small chunks (256 tokens) indexed for retrieval scoring. Extract principles from large chunks, link them to small chunks for precision. Extraction chunk size and retrieval chunk size are separate concerns.
v3 2026-04-05 Q Mapped to whitepaper sections
v2 2026-04-05 Q Imported SPEC-044 from model_specifications_v2.html
v1 2026-04-05 Q Created spec: SPEC-044: Hierarchical Chunking