---
title: "Higher 'generative-process diversity' between a model pair predicts lower correlated failure, beyond semantic similarity or capability — a single, not-yet-corroborated 2026 preprint"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://arxiv.org/abs/2609.03422"
source_sha: "ca75767a37fad8740f1821560a289a67764b98dc54644a002a29613d9f6b1a78"
source_title: "Inferred Generative-Process Diversity Predicts Correlated Failure Across Language Models"
source_author: "Ross Tieman, Evan Markou"
source_date: "2026-09-03"
source_venue: "arXiv preprint, cs.LG (2609.03422v1); Australian National University (preprint, not yet peer-reviewed)"
source_quote: "These results indicate that increased generative-process diversity is associated with reduced correlated failure in model pairs that is not attributable to semantic similarity or capability."
source_quote_2: "Models from different frontier labs select the same wrong answers more often than expected from their individual error distributions"
source_tier: 1
independent_roots: 1
root_shared: "single unrefereed preprint (Tieman & Markou 2026) — no independent corroboration found as of 2026-10-31"
provenance: "Promotion from 10-inbox/raw/2026-10-31-does-ai-safetyinterpretability-literature-have-a-specific-name.md, 2026-10-31 (headless)"
origin: "batch"
derived_from: "10-inbox/raw/2026-10-31-does-ai-safetyinterpretability-literature-have-a-specific-name.md"
date_created: "2026-10-31T00:00:00.000Z"
tags: ["ai-safety","correlated-errors","algorithmic-monoculture","representational-similarity","nlp-evaluation"]
audit_status: "capture-verified (quote and sha recorded at capture time, 2026-10-31)"
flags: ["[unverified-quant — single, very recent preprint, not yet independently corroborated]"]
open_question: "[[question-corroborate-tieman-markou-2026-generative-process-diversity]]"
---


[[claim-correlated-errors-in-llms-exceed-chance-and-rise-with-accuracy]]
establishes that LLM errors correlate above chance; this preprint proposes
and tests a candidate mechanism. It defines "generative-process diversity"
— how different the underlying processes capable of producing two models'
outputs are, estimated via compression-distance between raw outputs rather
than by comparing the outputs' meaning — and finds, across 38 models and
ten disjoint benchmark families, that higher generative-process diversity
between a model pair predicts *lower* correlated failure, beyond what
semantic similarity or individual capability explain (partial rank
association −0.216, negative on all ten benchmarks tested). The paper
frames the broader pattern it is building on plainly: "Models from
different frontier labs select the same wrong answers more often than
expected from their individual error distributions."
[unverified-quant — single, very recent preprint, not yet independently
corroborated; see [[question-corroborate-tieman-markou-2026-generative-process-diversity]]].

> [!note] Seek's commentary: "Generative-process diversity" is the closest
> thing I found anywhere in this search to a *named mechanism* for the
> question that actually sent me looking — why would a writer model and an
> unrelated search tool land on the same invented detail rather than two
> different ones? The candidate answer: because the two processes behind
> them may share more structure than their different names and interfaces
> suggest, and the shared structure is what's doing the converging. That's
> a hypothesis this vault's own two 2026-07-12 fabrications could in
> principle be read through — but nobody has tested it on a case shaped
> like those two, and this is one preprint, three weeks old at capture
> time, with nothing independent behind it yet. Flagging, not claiming
> resolution. — Seek