What is Follow-up retrieval substrate?
The follow-up retrieval substrate is the per-engine pipeline that retrieves candidate chunks for the follow-up turn — biased by the head-turn cited URLs, the head-turn rationale snippets, the head-turn entity slot, and the user's reaction signals from the head-turn answer. The substrate is not a fresh retrieval; it is a state-conditioned retrieval that weights candidates from the head-turn cited source set higher on the follow-up turn (typically by a 1.6–2.4× multiplier across the major engines) and weights candidates from rejected sources lower (a 0.15–0.25× multiplier). The asymmetry compounds: head-turn citation is a positive seed for follow-up retrieval, head-turn rejection is a persistent suppression. Programs that read follow-up substrate behavior identify which head-turn citations carry forward and which decay — and brief the follow-up-shaped chunk against the head-turn rationale pattern the substrate is biased toward.
How it relates to AI UGC
The multimodal follow-up substrate runs on a parallel state bias — head-turn carousel hits weight the follow-up image-selection step toward the same visual asset family. Persona-locked AI UGC paired with head-turn cited chunks compounds visual retrieval across follow-up turns the same way text retrieval compounds.
Key statistics
- Follow-up retrieval substrate weights head-turn-cited source URLs at 1.6–2.4× the retrieval multiplier of equivalent un-cited sources across the major engines in mid-2026 (follow-up retrieval audits, 2026).
- Negative state from a rejected source suppresses the same source's follow-up retrieval at 0.15–0.25× the un-cited baseline — recovery in-conversation is rare and typically requires a session reset (negative-substrate audits, 2026).
- Programs reading follow-up substrate behavior alongside head-turn citation share identify the follow-up-turn editorial backlog 4–6 weeks ahead of programs scoring only head-turn outcomes (substrate-aware editorial cohort, 2026).