liveBook · Manning

Synthesis

As explained in Retrieval Augmented Generation, The Foundational Ideas

Synthesis generates a novel answer by integrating information from multiple evidence sources rather than merely extracting text.

Synthesis generates a novel answer by combining evidence scattered across multiple retrieved documents. It addresses cases where no single passage contains the complete answer and extraction from the strongest passage misses relevant context. section 5.0 section 5.7

A practical synthesis architecture separates evidence isolation from evidence combination:

  • Encode each passage independently and in parallel, preserving separate representations and allowing processing across many documents. section 5.0 section 5.7
  • Fuse the passage representations in the decoder, which attends to all retrieved evidence while generating each answer step. section 5.7
  • Prefer retrieval recall over perfect ranking: relevant information only needs to appear somewhere in the top-k set, although larger sets increase decoding latency. section 5.7

Apple revenue growth

A financial analyst asks about Apple's revenue growth using passages on emerging-market iPhone sales, Services growth, and resilient logistics. A synthesizing reader should combine all three drivers rather than return only the first strong match. In practice, smaller general-purpose readers often remain extractive and let the strongest passage dominate, revealing that multi-document access alone does not guarantee synthesis. section 5.5

How Retrieval Augmented Generation, The Foundational Ideas covers it

How deep it goes
Explains independent encoding, decoder fusion, recall-oriented retrieval, and latency trade-offs, but defers production deployment details.
Examples
Uses an Apple revenue-growth question and contrasts multi-document synthesis with precise-fact retrieval among distractors.
What it assumes
Assumes familiarity with RAG, passage retrieval, and encoder-decoder models while clarifying their multi-document limitation.
What you can do after
Evaluate/understand when independent passage encoding and decoder fusion support synthesis across scattered evidence.
How it connects
Connects retrieval, RAG, encoder-decoder models, attention, and inference latency.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Retrieval Augmented Generation, The Foundational Ideas. The section links point to the parts of the book it draws on.