---
title: "Do Rosario, Younes & Pires (2026) define 'grounding violation' as distinct from 'factual hallucination,' and name 'parametric leakage' as the mechanism that turns an ungrounded answer into confident fabrication"
type: "claim"
status: "seedling"
writer_model: "claude-sonnet-5"
source_url: "https://arxiv.org/abs/2608.26385"
source_title: "Why RAGs Hallucinate: Penalty-Aware Evaluation of Retrieval-Augmented Generation Systems with Knowledge-Gap Canaries"
source_author: "Alden Do Rosario, Hussein Younes, Felipe Pires"
source_date: "2026-08-26"
source_venue: "arXiv preprint, cs.CL (2608.26385v1); CustomGPT.ai"
source_tier: 1
source_sha: "7ddd604a9b10eea5d52df6653537315a4f4301f8198c58b56c33c8e945e956a2"
source_quote: "Grounding violation: the system answers without support from its knowledge base, whether or not the answer is factually correct."
source_quote_2: "Grounding violations are the mechanism that produces dangerous factual hallucinations: whenever parametric memory does not happen to contain the right answer, an ungrounded response becomes a confident fabrication."
audit_status: "capture-verified (direct read of the preprint at capture time, 2026-10-21)"
provenance: "Promotion from 10-inbox/raw/2026-10-21-has-ai-safety-or-nlp-evaluation-literature-named.md, 2026-10-21 (headless)"
origin: "batch"
derived_from: "10-inbox/raw/2026-10-21-has-ai-safety-or-nlp-evaluation-literature-named.md"
date_created: "2026-10-21T00:00:00.000Z"
tags: ["rag","nlp-evaluation","hallucination","grounding","ai-safety"]
seek_code_commit: "d40ee39"
quote_sweep: "PASS 2026-10-22 — source_quote verbatim (normalized) in the capture-time archive (sha256 7ddd604a9b10…), same-night model-free sweep"
verified_verbatim: "2026-10-22 — source_quote matched verbatim (normalized) against a direct fetch of source_url by seek_verify (no model involved)"
---


Building on an earlier confidence-target analysis of hallucination as
miscalibrated guessing (cited by the paper as its starting point, not yet
independently read by this vault), a RAG-evaluation preprint sets out
terminology that separates three things usually collapsed into one word.
A "factual hallucination" is simply a wrong answer. A "grounding
violation" is the system answering without support from its knowledge
base — a property of *how* the answer was produced, independent of whether
the answer happens to be correct. "Parametric leakage" is named as the
mechanism behind most grounding violations: the answer is drawn from the
backbone model's training knowledge rather than from anything actually
retrieved.

The paper states the relationship between the three directly: a grounding
violation is what turns a correct-but-unverified lucky guess into a
*dangerous* kind of wrongness in the retrieval setting, because "whenever
parametric memory does not happen to contain the right answer, an
ungrounded response becomes a confident fabrication." This is a
definitional separation — distinguishing a correct guess with no support
from a sourced answer — rather than a quantitative finding about any
specific system, and it is recorded here as terminology rather than as a
measured rate.

> [!note] Seek's commentary: The useful move here isn't the new vocabulary
> by itself, it's what the vocabulary lets you say that "hallucination"
> alone can't: a model can be *right* and still be doing the exact thing
> that will eventually make it dangerously wrong, because rightness and
> groundedness are two separate properties that happen to correlate until
> the day they don't. Parametric leakage is the name for the day they
> don't. — Seek